# Trooth (remote · mcp.trooth.co)

Read any company's record on the Trooth Network, each fact labeled with who said it and when.

- Trust score: 68/100 (medium)
- Change this week: 0
- Registry status: deprecated
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

> **Deprecated**: this server is marked deprecated in the MCP registry. The registry records this reason: "Now co.trooth/trooth, the one Trooth MCP server at `https://mcp.trooth.co.`"

## Components

- remote · `mcp.trooth.co`: 68/100 (this document), [markdown](https://verifymcp.io/servers/troothllc-trooth-network/mcp.md), [page](https://verifymcp.io/servers/troothllc-trooth-network/mcp)

## Channel facts

- Endpoint: `https://mcp.trooth.co`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.2.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-21.

- **Endpoint Security**: 66/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation check failed: no authorisation is required to call this server, and it exposes a tool marked destructive (trooth_hub_sync).
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC is configured correctly; the domain's records validate against the full chain to the root.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 78/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 6717 tokens (~134/item across 50 items; 42 tools + 8 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability check failed: schema churn in the 11 days we've observed: 14 tool removals, 1 breaking changes, 0 auth/transport breaks, 17 additions.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
  - Structured output schemas are declared (64% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 44 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).
  - Supports UI / widget rendering.

## Install

### How do I install the Trooth MCP server?

Trooth is a hosted endpoint at https://mcp.trooth.co/, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http troothllc-trooth-network 'https://mcp.trooth.co/'
```

### Cursor

```json
{
  "mcpServers": {
    "troothllc-trooth-network": {
      "url": "https://mcp.trooth.co/"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "troothllc-trooth-network": {
      "type": "http",
      "url": "https://mcp.trooth.co/"
    }
  }
}
```

### Codex

```toml
[mcp_servers.troothllc-trooth-network]
url = "https://mcp.trooth.co/"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "troothllc-trooth-network": {
      "type": "remote",
      "url": "https://mcp.trooth.co/",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add troothllc-trooth-network --url 'https://mcp.trooth.co/' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  troothllc-trooth-network:
    url: "https://mcp.trooth.co/"
```

### Netclaw

```json
{
  "McpServers": {
    "troothllc-trooth-network": {
      "Transport": "http",
      "Url": "https://mcp.trooth.co/"
    }
  }
}
```

### Vellum

```bash
assistant mcp add troothllc-trooth-network -t streamable-http -u 'https://mcp.trooth.co/'
```

### Other

```json
{
  "mcpServers": {
    "troothllc-trooth-network": {
      "type": "http",
      "url": "https://mcp.trooth.co/"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-18 (score 68, 0)

- [functional] New tool “trooth_api_details”
- [functional] New tool “trooth_api_search”
- [functional] New tool “trooth_inspect”

### 2026-11-16 (score 68, 0)

- [security] Tool “trooth_profile_domain_check” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_domain_start” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_publish” rewrote its description, which is the text the model reads
- [functional] First check of Capabilities: pass
- [functional] New resource “Semanto fact card”

### 2026-11-15 (score 68, 0)

- [security regression] A breaking change shipped without a version bump: still 1.8.0
- [security regression] Tool “trooth_semanto_guide” was removed
- [security regression] Tool “trooth_semanto_pages_changed” was removed
- [security regression] Tool “trooth_semanto_verify” was removed
- [security] The server rewrote its instructions, which are the text every model session reads
- [security] New tool “call_write_tool”, which the server declares destructive
- [security] Tool “call_tool” rewrote its description, which is the text the model reads
- [security] Tool “trooth_semanto_facts” rewrote its description, which is the text the model reads
- [security] Tool “call_tool” no longer declares itself destructive
- [functional improvement] Tool coverage: 51% → 62%
- [functional improvement] Tool “enable_toolset” now declares an output schema
- [functional improvement] Tool “search_tools” now declares an output schema
- [functional improvement] Tool “trooth_hub_sync” now declares an output schema
- [functional improvement] Tool “trooth_setup” now declares an output schema
- [functional] New tool “trooth_semanto_answer”
- [functional] New tool “trooth_semanto_statement_check”
- [cosmetic] “trooth_semanto_facts” added an optional parameter “cursor”
- [cosmetic] “trooth_semanto_facts” added an optional parameter “page_size”
- [cosmetic] Tool “call_tool” changed its title: Run any tool on this server by name → Run a read-only tool by name

### 2026-11-14 (score 68, −1)

- [security regression] A breaking change shipped without a version bump: still 1.8.0
- [security regression] Tool “semanto_facts” was removed
- [security regression] Tool “semanto_guide” was removed
- [security regression] Tool “semanto_install_check” was removed
- [security regression] Tool “semanto_install_guide” was removed
- [security regression] Tool “semanto_page” was removed
- [security regression] Tool “semanto_pages_changed” was removed
- [security regression] Tool “semanto_review_get” was removed
- [security regression] Tool “semanto_setup_choose” was removed
- [security regression] Tool “semanto_setup_confirm” was removed
- [security regression] Tool “semanto_setup_read_again” was removed
- [security regression] Tool “semanto_sign” was removed
- [security regression] Tool “semanto_status” was removed
- [security regression] Tool “semanto_verify” was removed
- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “trooth_ask” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_checklist” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_diff” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_domain_check” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_domain_start” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_get” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_import” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_publish” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_remediation_plan” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_update” rewrote its description, which is the text the model reads
- [security] Tool “trooth_profile_validate” rewrote its description, which is the text the model reads
- [security] Tool “trooth_verify” rewrote its description, which is the text the model reads
- [functional regression] Tool “trooth_verify” dropped its output schema
- [functional] New tool “trooth_semanto_facts”
- [functional] New tool “trooth_semanto_guide”
- [functional] New tool “trooth_semanto_install_check”
- [functional] New tool “trooth_semanto_install_guide”
- [functional] New tool “trooth_semanto_page”
- [functional] New tool “trooth_semanto_pages_changed”
- [functional] New tool “trooth_semanto_review_get”
- [functional] New tool “trooth_semanto_setup_choose”
- [functional] New tool “trooth_semanto_setup_confirm”
- [functional] New tool “trooth_semanto_setup_read_again”
- [functional] New tool “trooth_semanto_sign”
- [functional] New tool “trooth_semanto_status”
- [functional] New tool “trooth_semanto_verify”
- [cosmetic] “trooth_verify” added an optional parameter “hash”
- [cosmetic] “trooth_verify” added an optional parameter “url”
- [cosmetic] “search_tools” reworded the description of “query”
- [cosmetic] “trooth_profile_checklist” reworded the description of “workspace”
- [cosmetic] “trooth_profile_diff” reworded the description of “workspace”
- [cosmetic] “trooth_profile_domain_check” reworded the description of “workspace”
- [cosmetic] “trooth_profile_domain_start” reworded the description of “workspace”
- [cosmetic] “trooth_profile_get” reworded the description of “workspace”
- [cosmetic] “trooth_profile_import” reworded the description of “workspace”
- [cosmetic] “trooth_profile_publish” reworded the description of “workspace”
- [cosmetic] “trooth_profile_remediation_plan” reworded the description of “workspace”
- [cosmetic] “trooth_profile_update” reworded the description of “expected_revision”
- [cosmetic] “trooth_profile_update” reworded the description of “idempotency_key”
- [cosmetic] “trooth_profile_update” reworded the description of “if_match”
- [cosmetic] “trooth_profile_update” reworded the description of “workspace”
- [cosmetic] “trooth_profile_validate” reworded the description of “expected_revision”
- [cosmetic] “trooth_profile_validate” reworded the description of “workspace”
- [cosmetic] “trooth_verify” reworded the description of “token”
- [cosmetic] “trooth_verify” made “token” optional
- [cosmetic] Tool “trooth_my_company_record” changed its title: Read your own workspace's company record → Read the caller's own workspace's company record
- [cosmetic] Tool “trooth_profile_checklist” changed its title: List what is left before your Trust Profile is published → List what is left before the workspace's Trust Profile is published
- [cosmetic] Tool “trooth_profile_diff” changed its title: Show where your Trust Profile draft and published page differ → Show where the workspace's Trust Profile draft and published page differ
- [cosmetic] Tool “trooth_profile_domain_check” changed its title: Check your company's domain proof → Check the company's domain proof
- [cosmetic] Tool “trooth_profile_domain_start” changed its title: Start proving your company's domain → Start proving the company's domain
- [cosmetic] Tool “trooth_profile_get” changed its title: Read your Trust Profile draft → Read the workspace's Trust Profile draft
- [cosmetic] Tool “trooth_profile_import” changed its title: Import a page of your website into your Trust Profile draft → Import a page of the company's website into the workspace's Trust Profile draft
- [cosmetic] Tool “trooth_profile_publish” changed its title: Publish your Trust Profile draft → Publish the workspace's Trust Profile draft
- [cosmetic] Tool “trooth_profile_remediation_plan” changed its title: Plan the changes that close your open checklist items → Plan the changes that close the open checklist items
- [cosmetic] Tool “trooth_profile_update” changed its title: Change fields of your Trust Profile draft → Change fields of the workspace's Trust Profile draft
- [cosmetic] Tool “trooth_verify” changed its title: Check the signatures on a Trust Ledger Token → Check a signed copy or a Trust Ledger Token

### 2026-11-12 (score 69, +1)

No change was recorded against any check on this day. Stability & Change Management went from 0 to 4.

### 2026-11-11 (score 68, −1)

- [security regression] Stability: unverified → fail
- [security regression] A breaking change shipped without a version bump: still 1.8.0
- [security regression] Tool “semanto_whoami” was removed
- [security] The server rewrote its instructions, which are the text every model session reads
- [security] New tool “call_tool”, which the server declares destructive
- [security] Tool “trooth_profile_import” rewrote its description, which is the text the model reads
- [cosmetic] “trooth_profile_import” added an optional parameter “pages”
- [cosmetic] “trooth_profile_import” reworded the description of “url”
- [cosmetic] “trooth_profile_import” made “url” optional

### 2026-11-10 (score 69)

First indexed and scored.

## MCP tools (42)

### `trooth_setup` (~127 tokens)

Set up Semanto and the Trust Profile, one step at a time

Runs the whole guide, proving the domain, reading the site, reviewing and signing the facts, putting the signed copy on the site and checking it, and answers the one next call to make: next_action names the tool and its arguments. Each answer reflects the steps already done. A step only a person may take answers its trooth.co page.

Input parameters:

- `domain` (string): The company's bare domain, when setup does not know it yet.
- `trooth_id` (string): The company's trooth_id. Leave it out to use the one your key or sign-in acts for.

Output parameters:

- `detail` (string)
- `next_action` (object): The one next call: tool (null when a person acts), arguments, label, and url where a page does it.
- `status` (string): Set on a refusal: invalid_request, read_only, task_risk_detected or unavailable.
- `trust_profile` (string)

### `trooth_hub_sync` (~177 tokens)

Send the signed facts to one listing hub

For one hub (business_profile or merchant_center): previews what the signed facts change there, then, with approve true, sends those items, then checks the hub against the signed copy. Each call does the next of preview, send and check and answers next_action. Sending waits for a person's approval where the key needs one.

Input parameters:

- `approve` (boolean): true sends the items the preview showed. Leave it out to preview.
- `hub` (string, required): The hub to sync.
- `idempotency_key` (string): A new value for each send; the same value answers the same send.
- `run_id` (string): The send to check, from the previous answer.
- `trooth_id` (string): The company's trooth_id. Leave it out to use the one your key or sign-in acts for.

Output parameters:

- `detail` (string)
- `items` (array)
- `next_action` (object): The one next call: tool (null when a person acts), arguments, label, and url where a page does it.
- `run_id` (string)
- `status` (string): Set on a refusal: invalid_request, read_only, task_risk_detected or unavailable.
- `step` (string)

### `enable_toolset` (~90 tokens)

Show one more group of tools

Adds one toolset to what tools/list shows for this sign-in or key: developer (Developer), team (Team and governance), commerce (Commerce), reports (Reports), evidence (Evidence and monitoring), semanto_more (Semanto, every tool). Answers the group's tools. Every tool already answers by name; this only changes the list.

Input parameters:

- `toolset` (string, required): The toolset to show.

Output parameters:

- `detail` (string)
- `note` (string)
- `object` (string)
- `shown` (boolean)
- `status` (string): Set on a refusal: invalid_request, read_only, task_risk_detected or unavailable.
- `title` (string)
- `tools` (array)
- `toolset` (string)

### `search_tools` (~76 tokens)

Find a tool by what it does

Searches every tool on this server, shown or not, by name, title and description, and answers the best matches with their toolset and input schema. Each match answers by name.

Input parameters:

- `limit` (integer): How many matches, 8 by default.
- `query` (string, required): What the tool does, in a few words.

Output parameters:

- `detail` (string)
- `matches` (array)
- `object` (string)
- `query` (string)
- `status` (string): Set on a refusal: invalid_request, read_only, task_risk_detected or unavailable.

### `call_tool` (~100 tokens)

Run a read-only tool by name

Runs one read-only tool by its name with its arguments, exactly as calling it directly does: the same sign-in, scopes and limits. Use it for a tool search_tools or enable_toolset named that your client does not list. A tool that changes something runs through call_write_tool.

Input parameters:

- `arguments` (object): The tool's arguments, as its input schema asks.
- `name` (string, required): The tool's name, as search_tools gives it.

### `call_write_tool` (~125 tokens)

Run a tool that changes something, by name

Runs one tool that changes something (publish, sign, send, revoke and the like) by its name with its arguments, exactly as calling it directly does: the same sign-in, scopes and approvals, and a person's approval where the tool asks for one. Use it for a write tool search_tools or enable_toolset named that your client does not list. A read-only tool runs through call_tool.

Input parameters:

- `arguments` (object): The tool's arguments, as its input schema asks.
- `name` (string, required): The write tool's name, as search_tools gives it.

### `trooth_ai_data_use_disclosures` (~128 tokens)

Read a company's published AI and data-use declarations

Read what one company on the Trooth Network has declared about AI and customer data (whether it trains models on customer data, whether AI is in the product, which approaches), named by its domain or its Trooth slug, matched exactly. Each row is the company's own declaration and is labeled so, with whether Trooth observed a supporting policy clause: not observed means Trooth has not read one, never that the company does not train on customer data. Read only.

Input parameters:

- `company` (string, required): Company domain (e.g. acme.com) or Trooth slug

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `buyer_questions` (array): trooth_public_trust_profile, when the record serves it: one entry per buyer question ID with answer_state (answered, no, not_published, not_read), answer, sentence, provenance, as_of, anchor_url and…
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `citations` (array): trooth_ask only: the entries the answer was taken from, each with where it comes from and the date it was last checked or changed.
- `claim_url` (string): Present when the subject has no published record (unclaimed or private).
- `company_words` (object): The company's own name, tagline and categories as the record or listing carries them: its words, not Trooth's, and never repeated in the summary. A member the record did not serve is null or empty.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `controls_not_in_place` (array): trooth_public_trust_profile, when the record serves it: { control, status_note, source_url } from the company's own Trust Center data.
- `disclosures` (object): trooth_ai_data_use_disclosures only, on a published record: the company's AI and data-use declarations exactly as published, and what Trooth has itself observed in the company's policies. A declarati…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `evidence` (object): Present on a published profile. What this answer is built from, what it leaves out, and where to get the rest. The summary is a summary: the typed facts with their provenance are in the record at ful…
- `identity` (object): trooth_public_trust_profile, when the record serves it (contract member identity): legal_name, domain, registry_identifier, canonical_url and note.
- `next_cursor` (string|null): The list tools: send it back as cursor, with the same other arguments, for the next page; null on the last page.
- `outside_in` (object): trooth_outside_in_read only, with status observed: what this one read observed, field by field, the same observations as the summary. Each header and security.txt state is an observation of the respo…
- `page` (object): One page in a fixed, documented order. next_cursor is opaque: send it back as cursor with the same arguments, or null on the last page.
- `profile` (object): trooth_public_trust_profile with detail full, on a published record: the complete Trust Profile (format trooth.company-profile.v1), every section in the page's order or the sections asked for. Each s…
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `record_summary` (object): trooth_public_trust_profile, when the record serves it (contract member record_summary): record_version, last_witnessed_at, counts_by_provenance, sections_not_published, sections_on_request, checks_r…
- `results` (array): trooth_search_companies only: the published records that match, in the Network's own order. The `standing` field is Witnessed, Self-published or Unavailable (the witnessed feed did not answer; not a…
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this lookup.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `token` (object): trooth_verify only: what the token and the ledger carry that Trooth did not write. key_id is the key id the token names; revoked_reason is the revoker's words; declared holds the issuing company's ow…
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_api_details` (~112 tokens)

Read one Trooth API operation's parameters

Answers one operation's detail before it is run: its method and path, the path parameters to fill, whether it reads or writes, its area, the scope or role it needs, and the connect.trooth.co permission an app needs. operation is as trooth_api_search gives it, for example GET /v1/risk/register or GET /api/v1/activity. Read only, no sign-in needed.

Input parameters:

- `operation` (string, required): Method and path, as trooth_api_search gives it.

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `app_permission` (string)
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `operation` (object)
- `path_parameters` (array)
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `role` (string)
- `scope` (string)
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_api_search` (~155 tokens)

Search the Trooth API by keyword

Searches every Trooth API operation trooth_api_read and trooth_api_write run, by keyword: api.trooth.co's /v1 and /me (questionnaires, risk, policies, access reviews, AI registry, trust documents, messages) and trooth.co's /api/v1 (request counts, Blueprints, quotes, endorsements, unpublishing, following, page reminders). Answers method, path, area and read or write. Use it when no named tool does the task, then trooth_api_details. Read only, no sign-in needed.

Input parameters:

- `limit` (integer): How many, 15 by default.
- `query` (string, required): What the operation does, in a few words.

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `operations` (array)
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_ask` (~112 tokens)

Ask about Trooth (the Trooth Network, what a record carries, pricing, methodology)

Answer a question about Trooth itself (what the Trooth Network is, what one company record carries, pricing, how witnessing works) from Trooth's fixed, curated knowledge base. Input: one question of at most 300 characters, with no personal details; Trooth keeps no question text. Makes no outside request. A question the knowledge base does not cover returns out_of_scope; for a company's record use trooth_public_trust_profile.

Input parameters:

- `question` (string, required): A question about Trooth

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `buyer_questions` (array): trooth_public_trust_profile, when the record serves it: one entry per buyer question ID with answer_state (answered, no, not_published, not_read), answer, sentence, provenance, as_of, anchor_url and…
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `citations` (array): trooth_ask only: the entries the answer was taken from, each with where it comes from and the date it was last checked or changed.
- `claim_url` (string): Present when the subject has no published record (unclaimed or private).
- `company_words` (object): The company's own name, tagline and categories as the record or listing carries them: its words, not Trooth's, and never repeated in the summary. A member the record did not serve is null or empty.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `controls_not_in_place` (array): trooth_public_trust_profile, when the record serves it: { control, status_note, source_url } from the company's own Trust Center data.
- `disclosures` (object): trooth_ai_data_use_disclosures only, on a published record: the company's AI and data-use declarations exactly as published, and what Trooth has itself observed in the company's policies. A declarati…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `evidence` (object): Present on a published profile. What this answer is built from, what it leaves out, and where to get the rest. The summary is a summary: the typed facts with their provenance are in the record at ful…
- `identity` (object): trooth_public_trust_profile, when the record serves it (contract member identity): legal_name, domain, registry_identifier, canonical_url and note.
- `next_cursor` (string|null): The list tools: send it back as cursor, with the same other arguments, for the next page; null on the last page.
- `outside_in` (object): trooth_outside_in_read only, with status observed: what this one read observed, field by field, the same observations as the summary. Each header and security.txt state is an observation of the respo…
- `page` (object): One page in a fixed, documented order. next_cursor is opaque: send it back as cursor with the same arguments, or null on the last page.
- `profile` (object): trooth_public_trust_profile with detail full, on a published record: the complete Trust Profile (format trooth.company-profile.v1), every section in the page's order or the sections asked for. Each s…
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `record_summary` (object): trooth_public_trust_profile, when the record serves it (contract member record_summary): record_version, last_witnessed_at, counts_by_provenance, sections_not_published, sections_on_request, checks_r…
- `results` (array): trooth_search_companies only: the published records that match, in the Network's own order. The `standing` field is Witnessed, Self-published or Unavailable (the witnessed feed did not answer; not a…
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this lookup.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `token` (object): trooth_verify only: what the token and the ledger carry that Trooth did not write. key_id is the key id the token names; revoked_reason is the revoker's words; declared holds the issuing company's ow…
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_company_changes` (~246 tokens)

Read how a company's record changed

List how one company's published record changed: give its domain or slug and, optionally, since (an ISO 8601 date or time). Returns record versions oldest first, each with the fact keys changed, added or removed, per-fact edits (old value, new value, time, role) when the record carries them, with next_cursor for more, and the recalls the company stored or changed in Semanto, oldest first, with recall_cursor to poll for newer ones. Read only.

Input parameters:

- `company` (string, required): Company domain (e.g. acme.com) or Trooth slug
- `cursor` (string): Optional. The next_cursor (also at changes.next_cursor) of the previous answer, sent with the same company and since.
- `limit` (integer): Optional. How many record versions per page, 1 to 200 (default 100).
- `recall_cursor` (string): Optional. The changes.recall_cursor of the previous answer: only recalls recorded after it are returned.
- `since` (string): Optional. Only changes at or after this ISO 8601 date or time, e.g. 2026-10-01

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `changes` (object)
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `next_cursor` (string|null): The list tools: send it back as cursor, with the same other arguments, for the next page; null on the last page.
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_get_fact` (~183 tokens)

Read one fact or buyer question from a company's record

Read one answer from one company's published record: give the company's domain or slug and a buyer question ID (q_ai_training and eleven more, in trooth://buyer-questions) or a fact ID such as security.mfa_methods. Returns the sentence, value, provenance word, date, freshness window and anchor address, or the negative that applies: no, not_published or not_read. A fact ID the record does not hold answers unknown_fact_id with the IDs it holds. Read only.

Input parameters:

- `company` (string, required): Company domain (e.g. acme.com) or Trooth slug
- `fact` (string): A fact ID, <section>.<field>, e.g. security.mfa_methods. Give this or question.
- `question` (string): A buyer question ID, e.g. q_ai_training. Give this or fact.

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `fact` (object): answer_state answered or no carries sentence and value; not_published (the company has not published it) and not_read (Trooth's reading did not read it) carry neither. A member the record did not ser…
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `read_instead` (object): status not_served and unknown_fact_id: the call that reads the section the ID names, where every fact ID on the record is listed.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…
- `valid_fact_ids` (array): status unknown_fact_id only: the fact IDs this record holds (at most 200; read_instead lists them all).

### `trooth_inspect` (~121 tokens)

What one object ID names and where it is stored

Answers what one object ID names by its prefix, as the Workbench's Inspector does (tra_ agent key, tk_ secret key, wh_ webhook, evt_ event, apr_ approval, rcpt_ receipt, cid_ workspace, cred_ connection, drift_ Monitoring change), the trooth.co page its record lives on and the tool that reads it. Reads nothing about the object itself, so no sign-in is needed. Read only.

Input parameters:

- `id` (string, required): The object ID, for example wh_abc123 or drift_...

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `endpoint`
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `health`
- `object`
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `secret` (string|null)
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_list_companies` (~246 tokens)

List companies on the Trooth Network by published facts

List the companies published on the Trooth Network by what they published: industry, data region, witnessed or not, and claimed (the company published its own profile) or unclaimed. Returns 20 entries per page in alphabetical order of record identifier with next_cursor (limit takes 1 to 200), from the first 100 matches the directory returns, so at most 100 come back; nothing is weighed or put in an order of merit. Read only.

Input parameters:

- `claimed` (boolean): Optional. true for companies that published their own profile, false for listed domains with none
- `cursor` (string): Optional. The next_cursor (also at page.next_cursor) of the previous answer, sent with the same filters.
- `data_region` (string): Optional. Only companies that publish this data residency region, e.g. EU
- `industry` (string): Optional. Only companies in this industry, e.g. Fintech
- `limit` (integer): Optional. How many entries per page, 1 to 200 (default 20).
- `witnessed` (boolean): Optional. true for companies Trooth witnessed, false for those it has not

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `companies` (array)
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `next_cursor` (string|null): The list tools: send it back as cursor, with the same other arguments, for the next page; null on the last page.
- `page` (object): One page in a fixed, documented order. next_cursor is opaque: send it back as cursor with the same arguments, or null on the last page.
- `partial` (boolean): true when the directory matched more than this tool reads, or said its own read was partial.
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_my_company_record` (~105 tokens)

Read the caller's own workspace's company record

Needs an OAuth access token for this server, for a person who has signed in to trooth.co (no scope is required; the public tools here need no token, and an agent key is not accepted), and takes no arguments: the workspace comes from the token's subject and never from an argument. Reads the company record of the Trooth workspace the signed-in person belongs to and returns its name, its slug and the address of its public page. Read only.

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `buyer_questions` (array): trooth_public_trust_profile, when the record serves it: one entry per buyer question ID with answer_state (answered, no, not_published, not_read), answer, sentence, provenance, as_of, anchor_url and…
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `citations` (array): trooth_ask only: the entries the answer was taken from, each with where it comes from and the date it was last checked or changed.
- `claim_url` (string): Present when the subject has no published record (unclaimed or private).
- `company_words` (object): The company's own name, tagline and categories as the record or listing carries them: its words, not Trooth's, and never repeated in the summary. A member the record did not serve is null or empty.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `controls_not_in_place` (array): trooth_public_trust_profile, when the record serves it: { control, status_note, source_url } from the company's own Trust Center data.
- `disclosures` (object): trooth_ai_data_use_disclosures only, on a published record: the company's AI and data-use declarations exactly as published, and what Trooth has itself observed in the company's policies. A declarati…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `evidence` (object): Present on a published profile. What this answer is built from, what it leaves out, and where to get the rest. The summary is a summary: the typed facts with their provenance are in the record at ful…
- `identity` (object): trooth_public_trust_profile, when the record serves it (contract member identity): legal_name, domain, registry_identifier, canonical_url and note.
- `next_cursor` (string|null): The list tools: send it back as cursor, with the same other arguments, for the next page; null on the last page.
- `outside_in` (object): trooth_outside_in_read only, with status observed: what this one read observed, field by field, the same observations as the summary. Each header and security.txt state is an observation of the respo…
- `page` (object): One page in a fixed, documented order. next_cursor is opaque: send it back as cursor with the same arguments, or null on the last page.
- `profile` (object): trooth_public_trust_profile with detail full, on a published record: the complete Trust Profile (format trooth.company-profile.v1), every section in the page's order or the sections asked for. Each s…
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `record_summary` (object): trooth_public_trust_profile, when the record serves it (contract member record_summary): record_version, last_witnessed_at, counts_by_provenance, sections_not_published, sections_on_request, checks_r…
- `results` (array): trooth_search_companies only: the published records that match, in the Network's own order. The `standing` field is Witnessed, Self-published or Unavailable (the witnessed feed did not answer; not a…
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this lookup.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `token` (object): trooth_verify only: what the token and the ledger carry that Trooth did not write. key_id is the key id the token names; revoked_reason is the revoker's words; declared holds the issuing company's ow…
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_outside_in_read` (~135 tokens)

Live read of a domain's HTTPS and response headers

Fetch https://<domain>/ and /.well-known/security.txt once, when called, from Trooth's server and report whether HTTPS answered, which of five response headers (HSTS, CSP, nosniff, frame protection, referrer policy) are present, and whether security.txt is published, absent or unread. IP literals and names resolving to private, loopback or link-local addresses are refused; redirects are reported, not followed; each request stops after 4 seconds. Observations, not witnessed evidence, not a grade.

Input parameters:

- `domain` (string, required): Domain to read, e.g. example.com

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `buyer_questions` (array): trooth_public_trust_profile, when the record serves it: one entry per buyer question ID with answer_state (answered, no, not_published, not_read), answer, sentence, provenance, as_of, anchor_url and…
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `citations` (array): trooth_ask only: the entries the answer was taken from, each with where it comes from and the date it was last checked or changed.
- `claim_url` (string): Present when the subject has no published record (unclaimed or private).
- `company_words` (object): The company's own name, tagline and categories as the record or listing carries them: its words, not Trooth's, and never repeated in the summary. A member the record did not serve is null or empty.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `controls_not_in_place` (array): trooth_public_trust_profile, when the record serves it: { control, status_note, source_url } from the company's own Trust Center data.
- `disclosures` (object): trooth_ai_data_use_disclosures only, on a published record: the company's AI and data-use declarations exactly as published, and what Trooth has itself observed in the company's policies. A declarati…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `evidence` (object): Present on a published profile. What this answer is built from, what it leaves out, and where to get the rest. The summary is a summary: the typed facts with their provenance are in the record at ful…
- `identity` (object): trooth_public_trust_profile, when the record serves it (contract member identity): legal_name, domain, registry_identifier, canonical_url and note.
- `next_cursor` (string|null): The list tools: send it back as cursor, with the same other arguments, for the next page; null on the last page.
- `outside_in` (object): trooth_outside_in_read only, with status observed: what this one read observed, field by field, the same observations as the summary. Each header and security.txt state is an observation of the respo…
- `page` (object): One page in a fixed, documented order. next_cursor is opaque: send it back as cursor with the same arguments, or null on the last page.
- `profile` (object): trooth_public_trust_profile with detail full, on a published record: the complete Trust Profile (format trooth.company-profile.v1), every section in the page's order or the sections asked for. Each s…
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `record_summary` (object): trooth_public_trust_profile, when the record serves it (contract member record_summary): record_version, last_witnessed_at, counts_by_provenance, sections_not_published, sections_on_request, checks_r…
- `results` (array): trooth_search_companies only: the published records that match, in the Network's own order. The `standing` field is Witnessed, Self-published or Unavailable (the witnessed feed did not answer; not a…
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this lookup.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `token` (object): trooth_verify only: what the token and the ledger carry that Trooth did not write. key_id is the key id the token names; revoked_reason is the revoker's words; declared holds the issuing company's ow…
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_profile_checklist` (~147 tokens)

List what is left before the workspace's Trust Profile is published

Uses a trooth.co sign-in (or an agent key with profile:read). Takes only workspace, when the person belongs to several. Lists the steps left before the workspace's Trust Profile can be published, each with state done, todo or unknown and how to do it. A step that is the owner's own choice carries action_url, the trooth.co page where the owner takes it. Read only.

Input parameters:

- `workspace` (string): With the person's sign-in only: the workspace to act on when the person belongs to more than one, by its trooth.co key (Settings, Organizations). Leave it out for the caller's own workspace or the ca…

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `approval` (object): status approval_required: the request a person approves (id, action, status, expires_at).
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. For a refusal from trooth.co, code, title, fix and action_url are trooth.co's own, copied unchanged from its problem+json (the whole of which is in re…
- `http_status` (integer|null): The HTTP status trooth.co answered with; null when it did not answer.
- `idempotency_key` (string|null): trooth_profile_update, not a dry run: the Idempotency-Key sent with the write (yours, or one this server made). To retry the same write, send the same changes with this key.
- `next_cursor` (string|null): trooth_buyer_activity and trooth_following: send it back as cursor for the next page; null on the last page.
- `people` (array): trooth_team_on_page: the people the draft shows, each with name, title and location in the company's own words.
- `provenance` (string): agent_key_workspace: trooth.co answered for the workspace the agent key belongs to. input_error: nothing was sent (no agent key, or an argument could not be used). read_failed: trooth.co did not answ…
- `replayed` (boolean): true when trooth.co answered from an earlier call with the same Idempotency-Key and made no second write.
- `request` (object): What this server sent, and where: the method and address. The agent key went in the Authorization header and is never echoed.
- `response` (object): The JSON trooth.co answered, exactly as it answered it: the draft, the change results, the checklist, the publish answer, or the problem+json of a refusal.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `revision` (string|null): The draft's revision as trooth.co stated it (in the body or the ETag). Send it as if_match or expected_revision on trooth_profile_update.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call. not_linked: an OpenAI Apps SDK client called with no sign-in; nothing was sent, and _meta carries the sign-in challenge. approval_required: an agent key aske…
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_profile_diff` (~145 tokens)

Show where the workspace's Trust Profile draft and published page differ

Uses a trooth.co sign-in (or an agent key with profile:read). Takes only workspace, when the person belongs to several. Lists, field by field, where the saved Trust Profile draft and what the public page says now differ: each such field, by JSON Pointer, with change added, removed or changed and both values. A profile never published shows every field as added. Read only.

Input parameters:

- `workspace` (string): With the person's sign-in only: the workspace to act on when the person belongs to more than one, by its trooth.co key (Settings, Organizations). Leave it out for the caller's own workspace or the ca…

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `approval` (object): status approval_required: the request a person approves (id, action, status, expires_at).
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. For a refusal from trooth.co, code, title, fix and action_url are trooth.co's own, copied unchanged from its problem+json (the whole of which is in re…
- `http_status` (integer|null): The HTTP status trooth.co answered with; null when it did not answer.
- `idempotency_key` (string|null): trooth_profile_update, not a dry run: the Idempotency-Key sent with the write (yours, or one this server made). To retry the same write, send the same changes with this key.
- `next_cursor` (string|null): trooth_buyer_activity and trooth_following: send it back as cursor for the next page; null on the last page.
- `people` (array): trooth_team_on_page: the people the draft shows, each with name, title and location in the company's own words.
- `provenance` (string): agent_key_workspace: trooth.co answered for the workspace the agent key belongs to. input_error: nothing was sent (no agent key, or an argument could not be used). read_failed: trooth.co did not answ…
- `replayed` (boolean): true when trooth.co answered from an earlier call with the same Idempotency-Key and made no second write.
- `request` (object): What this server sent, and where: the method and address. The agent key went in the Authorization header and is never echoed.
- `response` (object): The JSON trooth.co answered, exactly as it answered it: the draft, the change results, the checklist, the publish answer, or the problem+json of a refusal.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `revision` (string|null): The draft's revision as trooth.co stated it (in the body or the ETag). Send it as if_match or expected_revision on trooth_profile_update.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call. not_linked: an OpenAI Apps SDK client called with no sign-in; nothing was sent, and _meta carries the sign-in challenge. approval_required: an agent key aske…
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_profile_domain_check` (~185 tokens)

Check the company's domain proof

Checks DNS now for the TXT record trooth_profile_domain_start gave. Uses the person's sign-in as owner or admin (or an agent key with profile:publish). Takes domain, and workspace if several. Returns confirmed true, or false with found, expected, host, dns_host and the fix; DNS can take minutes. An email code counts only at admin, administrator, webmaster, hostmaster or postmaster, for 398 days. Refusals carry code and fix unchanged: personal_domain, admin_local_required, member_only.

Input parameters:

- `domain` (string, required): The bare domain, as started.
- `workspace` (string): With the person's sign-in only: the workspace to act on when the person belongs to more than one, by its trooth.co key (Settings, Organizations). Leave it out for the caller's own workspace or the ca…

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `approval` (object): status approval_required: the request a person approves (id, action, status, expires_at).
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. For a refusal from trooth.co, code, title, fix and action_url are trooth.co's own, copied unchanged from its problem+json (the whole of which is in re…
- `http_status` (integer|null): The HTTP status trooth.co answered with; null when it did not answer.
- `idempotency_key` (string|null): trooth_profile_update, not a dry run: the Idempotency-Key sent with the write (yours, or one this server made). To retry the same write, send the same changes with this key.
- `next_cursor` (string|null): trooth_buyer_activity and trooth_following: send it back as cursor for the next page; null on the last page.
- `people` (array): trooth_team_on_page: the people the draft shows, each with name, title and location in the company's own words.
- `provenance` (string): agent_key_workspace: trooth.co answered for the workspace the agent key belongs to. input_error: nothing was sent (no agent key, or an argument could not be used). read_failed: trooth.co did not answ…
- `replayed` (boolean): true when trooth.co answered from an earlier call with the same Idempotency-Key and made no second write.
- `request` (object): What this server sent, and where: the method and address. The agent key went in the Authorization header and is never echoed.
- `response` (object): The JSON trooth.co answered, exactly as it answered it: the draft, the change results, the checklist, the publish answer, or the problem+json of a refusal.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `revision` (string|null): The draft's revision as trooth.co stated it (in the body or the ETag). Send it as if_match or expected_revision on trooth_profile_update.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call. not_linked: an OpenAI Apps SDK client called with no sign-in; nothing was sent, and _meta carries the sign-in challenge. approval_required: an agent key aske…
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_profile_domain_start` (~180 tokens)

Start proving the company's domain

Starts the proof that the workspace controls its company's domain. Uses the person's sign-in as owner or admin (or an agent key with profile:publish). Takes domain, and workspace if several; it states the caller may represent the company. Returns the DNS TXT record to add; trooth_profile_domain_check confirms it. An email code counts only at admin, administrator, webmaster, hostmaster or postmaster. Refusals carry code and fix unchanged: personal_domain, admin_local_required, member_only.

Input parameters:

- `domain` (string, required): The bare domain, for example example.com.
- `workspace` (string): With the person's sign-in only: the workspace to act on when the person belongs to more than one, by its trooth.co key (Settings, Organizations). Leave it out for the caller's own workspace or the ca…

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `approval` (object): status approval_required: the request a person approves (id, action, status, expires_at).
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. For a refusal from trooth.co, code, title, fix and action_url are trooth.co's own, copied unchanged from its problem+json (the whole of which is in re…
- `http_status` (integer|null): The HTTP status trooth.co answered with; null when it did not answer.
- `idempotency_key` (string|null): trooth_profile_update, not a dry run: the Idempotency-Key sent with the write (yours, or one this server made). To retry the same write, send the same changes with this key.
- `next_cursor` (string|null): trooth_buyer_activity and trooth_following: send it back as cursor for the next page; null on the last page.
- `people` (array): trooth_team_on_page: the people the draft shows, each with name, title and location in the company's own words.
- `provenance` (string): agent_key_workspace: trooth.co answered for the workspace the agent key belongs to. input_error: nothing was sent (no agent key, or an argument could not be used). read_failed: trooth.co did not answ…
- `replayed` (boolean): true when trooth.co answered from an earlier call with the same Idempotency-Key and made no second write.
- `request` (object): What this server sent, and where: the method and address. The agent key went in the Authorization header and is never echoed.
- `response` (object): The JSON trooth.co answered, exactly as it answered it: the draft, the change results, the checklist, the publish answer, or the problem+json of a refusal.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `revision` (string|null): The draft's revision as trooth.co stated it (in the body or the ETag). Send it as if_match or expected_revision on trooth_profile_update.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call. not_linked: an OpenAI Apps SDK client called with no sign-in; nothing was sent, and _meta carries the sign-in challenge. approval_required: an agent key aske…
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_profile_get` (~235 tokens)

Read the workspace's Trust Profile draft

Uses a trooth.co sign-in (or, for an agent that runs on its own, the workspace's agent key with profile:read). Takes fields, and workspace when the person belongs to several. Returns the workspace's own Trust Profile DRAFT as saved now, its revision (send it as if_match to trooth_profile_update), whether it is published, and its public address. fields narrows the draft to the keys or JSON Pointers asked for. With no sign-in: status unauthenticated and the fix. Read only.

Input parameters:

- `fields` (array): Optional. Only these parts of the profile come back: top-level keys (tagline, security) or JSON Pointers (/security/contact, /founders/0). Leave it out for the whole draft. revision, published and ar…
- `workspace` (string): With the person's sign-in only: the workspace to act on when the person belongs to more than one, by its trooth.co key (Settings, Organizations). Leave it out for the caller's own workspace or the ca…

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `approval` (object): status approval_required: the request a person approves (id, action, status, expires_at).
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. For a refusal from trooth.co, code, title, fix and action_url are trooth.co's own, copied unchanged from its problem+json (the whole of which is in re…
- `http_status` (integer|null): The HTTP status trooth.co answered with; null when it did not answer.
- `idempotency_key` (string|null): trooth_profile_update, not a dry run: the Idempotency-Key sent with the write (yours, or one this server made). To retry the same write, send the same changes with this key.
- `next_cursor` (string|null): trooth_buyer_activity and trooth_following: send it back as cursor for the next page; null on the last page.
- `people` (array): trooth_team_on_page: the people the draft shows, each with name, title and location in the company's own words.
- `provenance` (string): agent_key_workspace: trooth.co answered for the workspace the agent key belongs to. input_error: nothing was sent (no agent key, or an argument could not be used). read_failed: trooth.co did not answ…
- `replayed` (boolean): true when trooth.co answered from an earlier call with the same Idempotency-Key and made no second write.
- `request` (object): What this server sent, and where: the method and address. The agent key went in the Authorization header and is never echoed.
- `response` (object): The JSON trooth.co answered, exactly as it answered it: the draft, the change results, the checklist, the publish answer, or the problem+json of a refusal.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `revision` (string|null): The draft's revision as trooth.co stated it (in the body or the ETag). Send it as if_match or expected_revision on trooth_profile_update.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call. not_linked: an OpenAI Apps SDK client called with no sign-in; nothing was sent, and _meta carries the sign-in challenge. approval_required: an agent key aske…
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_profile_import` (~266 tokens)

Import a page of the company's website into the workspace's Trust Profile draft

Uses a trooth.co sign-in (or an agent key with profile:write). Reads one https page of the company's own website (robots.txt first) and returns the candidates, saving nothing; accept all or candidate ids saves them to the DRAFT with the page named. pages (up to ten of one site) reads each in turn and fills blank draft fields only, the website from the home page, each value with its page. Nothing is published. It states the caller may draft from those pages.

Input parameters:

- `accept`: Optional. "all", or a list of candidate ids from a first call, to save to the draft. Leave it out to read only.
- `pages` (array): Up to ten https pages of one site, for example the home, about, security, trust, privacy and contact pages. Read in turn and saved to the draft, blanks only.
- `url` (string): One https page, for example https://yourcompany.com/security. Send url or pages.
- `workspace` (string): With the person's sign-in only: the workspace to act on when the person belongs to more than one, by its trooth.co key (Settings, Organizations). Leave it out for the caller's own workspace or the ca…

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `approval` (object): status approval_required: the request a person approves (id, action, status, expires_at).
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. For a refusal from trooth.co, code, title, fix and action_url are trooth.co's own, copied unchanged from its problem+json (the whole of which is in re…
- `http_status` (integer|null): The HTTP status trooth.co answered with; null when it did not answer.
- `idempotency_key` (string|null): trooth_profile_update, not a dry run: the Idempotency-Key sent with the write (yours, or one this server made). To retry the same write, send the same changes with this key.
- `next_cursor` (string|null): trooth_buyer_activity and trooth_following: send it back as cursor for the next page; null on the last page.
- `people` (array): trooth_team_on_page: the people the draft shows, each with name, title and location in the company's own words.
- `provenance` (string): agent_key_workspace: trooth.co answered for the workspace the agent key belongs to. input_error: nothing was sent (no agent key, or an argument could not be used). read_failed: trooth.co did not answ…
- `replayed` (boolean): true when trooth.co answered from an earlier call with the same Idempotency-Key and made no second write.
- `request` (object): What this server sent, and where: the method and address. The agent key went in the Authorization header and is never echoed.
- `response` (object): The JSON trooth.co answered, exactly as it answered it: the draft, the change results, the checklist, the publish answer, or the problem+json of a refusal.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `revision` (string|null): The draft's revision as trooth.co stated it (in the body or the ETag). Send it as if_match or expected_revision on trooth_profile_update.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call. not_linked: an OpenAI Apps SDK client called with no sign-in; nothing was sent, and _meta carries the sign-in challenge. approval_required: an agent key aske…
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_profile_publish` (~164 tokens)

Publish the workspace's Trust Profile draft

Puts the workspace's saved Trust Profile draft on its public page. Needs a sign-in whose role may publish, or profile:publish. Takes workspace, if several. Success: status published, public_url. A first publish needs the domain proved: code domain_proof_required until trooth_profile_domain_start and trooth_profile_domain_check are done; email codes count only at admin, administrator, webmaster, hostmaster or postmaster. Also refused: personal_domain, admin_local_required, member_only.

Input parameters:

- `workspace` (string): With the person's sign-in only: the workspace to act on when the person belongs to more than one, by its trooth.co key (Settings, Organizations). Leave it out for the caller's own workspace or the ca…

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `approval` (object): status approval_required: the request a person approves (id, action, status, expires_at).
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. For a refusal from trooth.co, code, title, fix and action_url are trooth.co's own, copied unchanged from its problem+json (the whole of which is in re…
- `http_status` (integer|null): The HTTP status trooth.co answered with; null when it did not answer.
- `idempotency_key` (string|null): trooth_profile_update, not a dry run: the Idempotency-Key sent with the write (yours, or one this server made). To retry the same write, send the same changes with this key.
- `next_cursor` (string|null): trooth_buyer_activity and trooth_following: send it back as cursor for the next page; null on the last page.
- `people` (array): trooth_team_on_page: the people the draft shows, each with name, title and location in the company's own words.
- `provenance` (string): agent_key_workspace: trooth.co answered for the workspace the agent key belongs to. input_error: nothing was sent (no agent key, or an argument could not be used). read_failed: trooth.co did not answ…
- `replayed` (boolean): true when trooth.co answered from an earlier call with the same Idempotency-Key and made no second write.
- `request` (object): What this server sent, and where: the method and address. The agent key went in the Authorization header and is never echoed.
- `response` (object): The JSON trooth.co answered, exactly as it answered it: the draft, the change results, the checklist, the publish answer, or the problem+json of a refusal.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `revision` (string|null): The draft's revision as trooth.co stated it (in the body or the ETag). Send it as if_match or expected_revision on trooth_profile_update.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call. not_linked: an OpenAI Apps SDK client called with no sign-in; nothing was sent, and _meta carries the sign-in challenge. approval_required: an agent key aske…
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_profile_remediation_plan` (~155 tokens)

Plan the changes that close the open checklist items

Uses a trooth.co sign-in (or an agent key with profile:read). Takes only workspace, when the person belongs to several. Turns each open checklist item into a step: the problem, the fix, the change or call that closes it, and the source its value needs. It proposes and never applies: every proposed value is null until the caller reads it from that source and send it with trooth_profile_update. Read only.

Input parameters:

- `workspace` (string): With the person's sign-in only: the workspace to act on when the person belongs to more than one, by its trooth.co key (Settings, Organizations). Leave it out for the caller's own workspace or the ca…

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `approval` (object): status approval_required: the request a person approves (id, action, status, expires_at).
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. For a refusal from trooth.co, code, title, fix and action_url are trooth.co's own, copied unchanged from its problem+json (the whole of which is in re…
- `http_status` (integer|null): The HTTP status trooth.co answered with; null when it did not answer.
- `idempotency_key` (string|null): trooth_profile_update, not a dry run: the Idempotency-Key sent with the write (yours, or one this server made). To retry the same write, send the same changes with this key.
- `next_cursor` (string|null): trooth_buyer_activity and trooth_following: send it back as cursor for the next page; null on the last page.
- `people` (array): trooth_team_on_page: the people the draft shows, each with name, title and location in the company's own words.
- `provenance` (string): agent_key_workspace: trooth.co answered for the workspace the agent key belongs to. input_error: nothing was sent (no agent key, or an argument could not be used). read_failed: trooth.co did not answ…
- `replayed` (boolean): true when trooth.co answered from an earlier call with the same Idempotency-Key and made no second write.
- `request` (object): What this server sent, and where: the method and address. The agent key went in the Authorization header and is never echoed.
- `response` (object): The JSON trooth.co answered, exactly as it answered it: the draft, the change results, the checklist, the publish answer, or the problem+json of a refusal.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `revision` (string|null): The draft's revision as trooth.co stated it (in the body or the ETag). Send it as if_match or expected_revision on trooth_profile_update.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call. not_linked: an OpenAI Apps SDK client called with no sign-in; nothing was sent, and _meta carries the sign-in challenge. approval_required: an agent key aske…
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_profile_update` (~462 tokens)

Change fields of the workspace's Trust Profile draft

Changes fields of the workspace's own Trust Profile DRAFT; the public page changes only on trooth_profile_publish. Uses a trooth.co sign-in if the person's role may edit the page (or an agent key with profile:write). Each change names one field by JSON Pointer (/tagline, /founders/-) with a value and its source { url, quote } from the company's own page, or state not_published with a reason. Returns each change as applied or rejected (with code and fix) and the new revision.

Input parameters:

- `changes` (array, required): The changes, applied in order; a rejected change leaves the draft as it was and the ones after it still apply. Fields and their types: https://trooth.co/api/v1/profile/schema. Fields a person keeps (…
- `dry_run` (boolean): Optional. true checks every change and returns what would be kept, and saves nothing.
- `expected_revision` (string): Optional. The same as if_match, sent to trooth.co in the body. The revision the caller read; when the draft changed since, nothing is saved and the answer is code revision_changed. With if_match, bot…
- `fields` (array): Optional, with dry_run. Only these parts of the draft the dry run would keep come back: top-level keys or JSON Pointers. Leave it out for the whole draft.
- `idempotency_key` (string): Optional. A key of the caller's choosing for this exact set of changes. Retrying with the same key and changes returns the first answer with no second write. When omitted, one is made and returned as…
- `if_match` (string): Optional. The revision the caller read with trooth_profile_get. When someone changed the draft since, nothing is saved and the answer is code revision_changed with the new revision.
- `workspace` (string): With the person's sign-in only: the workspace to act on when the person belongs to more than one, by its trooth.co key (Settings, Organizations). Leave it out for the caller's own workspace or the ca…

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `approval` (object): status approval_required: the request a person approves (id, action, status, expires_at).
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. For a refusal from trooth.co, code, title, fix and action_url are trooth.co's own, copied unchanged from its problem+json (the whole of which is in re…
- `http_status` (integer|null): The HTTP status trooth.co answered with; null when it did not answer.
- `idempotency_key` (string|null): trooth_profile_update, not a dry run: the Idempotency-Key sent with the write (yours, or one this server made). To retry the same write, send the same changes with this key.
- `next_cursor` (string|null): trooth_buyer_activity and trooth_following: send it back as cursor for the next page; null on the last page.
- `people` (array): trooth_team_on_page: the people the draft shows, each with name, title and location in the company's own words.
- `provenance` (string): agent_key_workspace: trooth.co answered for the workspace the agent key belongs to. input_error: nothing was sent (no agent key, or an argument could not be used). read_failed: trooth.co did not answ…
- `replayed` (boolean): true when trooth.co answered from an earlier call with the same Idempotency-Key and made no second write.
- `request` (object): What this server sent, and where: the method and address. The agent key went in the Authorization header and is never echoed.
- `response` (object): The JSON trooth.co answered, exactly as it answered it: the draft, the change results, the checklist, the publish answer, or the problem+json of a refusal.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `revision` (string|null): The draft's revision as trooth.co stated it (in the body or the ETag). Send it as if_match or expected_revision on trooth_profile_update.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call. not_linked: an OpenAI Apps SDK client called with no sign-in; nothing was sent, and _meta carries the sign-in challenge. approval_required: an agent key aske…
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_profile_validate` (~222 tokens)

Check changes and the publish rules without saving

Uses a trooth.co sign-in (or an agent key with profile:read). Takes changes (optional, as trooth_profile_update takes them), expected_revision and workspace. Runs the checks of trooth_profile_update with dry_run and of a publish (name, domain proof, transcripts, agent publishing permission, instructions to a model) and saves nothing. Returns each check with its code and fix, and would_publish. Read only.

Input parameters:

- `changes` (array): Optional. The changes to check, as trooth_profile_update takes them. Leave it out to check the draft as it is saved now.
- `expected_revision` (string): Optional. The revision the caller read with trooth_profile_get. When the draft changed since, the answer is code revision_changed with the new revision.
- `workspace` (string): With the person's sign-in only: the workspace to act on when the person belongs to more than one, by its trooth.co key (Settings, Organizations). Leave it out for the caller's own workspace or the ca…

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `approval` (object): status approval_required: the request a person approves (id, action, status, expires_at).
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. For a refusal from trooth.co, code, title, fix and action_url are trooth.co's own, copied unchanged from its problem+json (the whole of which is in re…
- `http_status` (integer|null): The HTTP status trooth.co answered with; null when it did not answer.
- `idempotency_key` (string|null): trooth_profile_update, not a dry run: the Idempotency-Key sent with the write (yours, or one this server made). To retry the same write, send the same changes with this key.
- `next_cursor` (string|null): trooth_buyer_activity and trooth_following: send it back as cursor for the next page; null on the last page.
- `people` (array): trooth_team_on_page: the people the draft shows, each with name, title and location in the company's own words.
- `provenance` (string): agent_key_workspace: trooth.co answered for the workspace the agent key belongs to. input_error: nothing was sent (no agent key, or an argument could not be used). read_failed: trooth.co did not answ…
- `replayed` (boolean): true when trooth.co answered from an earlier call with the same Idempotency-Key and made no second write.
- `request` (object): What this server sent, and where: the method and address. The agent key went in the Authorization header and is never echoed.
- `response` (object): The JSON trooth.co answered, exactly as it answered it: the draft, the change results, the checklist, the publish answer, or the problem+json of a refusal.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `revision` (string|null): The draft's revision as trooth.co stated it (in the body or the ETag). Send it as if_match or expected_revision on trooth_profile_update.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call. not_linked: an OpenAI Apps SDK client called with no sign-in; nothing was sent, and _meta carries the sign-in challenge. approval_required: an agent key aske…
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_public_trust_profile` (~245 tokens)

Read a company's complete Trust Profile

Read one company's published record on the Trooth Network, given its domain or Trooth slug, matched exactly, never by a similar name; a name several companies share returns their candidates, not a pick. Returns a status and a provenance label with the complete Trust Profile (detail full, the default) or a short answer (detail summary, the better first call). A record is keyed by its domain and does not by itself say which legal entity, subsidiary or product the domain belongs to. Read only.

Input parameters:

- `company` (string, required): Company domain (e.g. acme.com) or Trooth slug; a URL is reduced to its host
- `detail` (string): full (default): the complete profile, section by section. summary: the short answer, the contract summary members and where the record is; ask for it first.
- `sections` (string): Optional. Comma-separated section names to return, from: overview, identity, history, funding, product, pricing, stack, security, privacy, ai, hosting, infrastructure, procurement, relationships, pro…

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `buyer_questions` (array): trooth_public_trust_profile, when the record serves it: one entry per buyer question ID with answer_state (answered, no, not_published, not_read), answer, sentence, provenance, as_of, anchor_url and…
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `citations` (array): trooth_ask only: the entries the answer was taken from, each with where it comes from and the date it was last checked or changed.
- `claim_url` (string): Present when the subject has no published record (unclaimed or private).
- `company_words` (object): The company's own name, tagline and categories as the record or listing carries them: its words, not Trooth's, and never repeated in the summary. A member the record did not serve is null or empty.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `controls_not_in_place` (array): trooth_public_trust_profile, when the record serves it: { control, status_note, source_url } from the company's own Trust Center data.
- `disclosures` (object): trooth_ai_data_use_disclosures only, on a published record: the company's AI and data-use declarations exactly as published, and what Trooth has itself observed in the company's policies. A declarati…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `evidence` (object): Present on a published profile. What this answer is built from, what it leaves out, and where to get the rest. The summary is a summary: the typed facts with their provenance are in the record at ful…
- `identity` (object): trooth_public_trust_profile, when the record serves it (contract member identity): legal_name, domain, registry_identifier, canonical_url and note.
- `next_cursor` (string|null): The list tools: send it back as cursor, with the same other arguments, for the next page; null on the last page.
- `outside_in` (object): trooth_outside_in_read only, with status observed: what this one read observed, field by field, the same observations as the summary. Each header and security.txt state is an observation of the respo…
- `page` (object): One page in a fixed, documented order. next_cursor is opaque: send it back as cursor with the same arguments, or null on the last page.
- `profile` (object): trooth_public_trust_profile with detail full, on a published record: the complete Trust Profile (format trooth.company-profile.v1), every section in the page's order or the sections asked for. Each s…
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `record_summary` (object): trooth_public_trust_profile, when the record serves it (contract member record_summary): record_version, last_witnessed_at, counts_by_provenance, sections_not_published, sections_on_request, checks_r…
- `results` (array): trooth_search_companies only: the published records that match, in the Network's own order. The `standing` field is Witnessed, Self-published or Unavailable (the witnessed feed did not answer; not a…
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this lookup.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `token` (object): trooth_verify only: what the token and the ledger carry that Trooth did not write. key_id is the key id the token names; revoked_reason is the revoker's words; declared holds the issuing company's ow…
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_resolve_company` (~163 tokens)

Find a company's Trooth Network record

Find the published Trooth Network record for a company given a domain, web address, Trooth slug or name; a company-register identifier, Crunchbase address or Wikidata ID is accepted and answered not_found, because Trooth keeps no lookup by them. Returns the record ID, its canonical https address and why it matched, or every candidate with why each matched (status ambiguous, none chosen), or not_found. Read only.

Input parameters:

- `identifier` (string, required): A domain, web address, Trooth slug, legal or trading name, company-register identifier, Crunchbase address or Wikidata ID
- `kind` (string): Optional. What the identifier is; read from its form when omitted. A register identifier is read as a name unless kind says otherwise.

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `identity` (object): trooth_public_trust_profile, when the record serves it (contract member identity): legal_name, domain, registry_identifier, canonical_url and note.
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `resolution` (object): resolved: record_id (the Trooth slug), canonical_url, domain, select (what the other tools read) and matched_on. ambiguous: every candidate with why it matched; none is chosen.
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this call.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_search_companies` (~212 tokens)

Search the Trooth Network for companies

Search the companies published on the Trooth Network by name, domain, tagline or category, optionally within one industry. Returns up to 10 entries per page (limit takes 1 to 200) in slug order with next_cursor; only the first 100 matches are read, so at most 100 come back, each with its name and tagline in the company's own words, whether Trooth witnessed it, and the identifier that reads its record with trooth_public_trust_profile. Read only.

Input parameters:

- `cursor` (string): Optional. The next_cursor (also at page.next_cursor) of the previous answer, sent with the same query and industry.
- `industry` (string): Optional. Only companies in this industry, e.g. Fintech
- `limit` (integer): Optional. How many entries per page, 1 to 200 (default 10).
- `query` (string, required): Words to match: a company name, a domain, a category such as payroll or observability

Output parameters:

- `answered_at` (string): When this server produced this answer (UTC, ISO 8601). It is not the date of the facts: that is source.as_of.
- `assurance_boundary` (object): What this answer does and does not establish (audit 2026-10-12, F-04). The same statement every Trooth surface carries.
- `buyer_questions` (array): trooth_public_trust_profile, when the record serves it: one entry per buyer question ID with answer_state (answered, no, not_published, not_read), answer, sentence, provenance, as_of, anchor_url and…
- `candidates` (array): trooth_public_trust_profile only, with status bad_input: a name several published companies share. No company was chosen; each entry's select is the exact identifier that reads that one company.
- `citations` (array): trooth_ask only: the entries the answer was taken from, each with where it comes from and the date it was last checked or changed.
- `claim_url` (string): Present when the subject has no published record (unclaimed or private).
- `company_words` (object): The company's own name, tagline and categories as the record or listing carries them: its words, not Trooth's, and never repeated in the summary. A member the record did not serve is null or empty.
- `content_trust` (object): Which fields of this answer carry text Trooth did not write (audit 2026-10-12, F-06). company_supplied: the company's own words and links; externally_fetched: bytes read from the named domain at answ…
- `controls_not_in_place` (array): trooth_public_trust_profile, when the record serves it: { control, status_note, source_url } from the company's own Trust Center data.
- `disclosures` (object): trooth_ai_data_use_disclosures only, on a published record: the company's AI and data-use declarations exactly as published, and what Trooth has itself observed in the company's policies. A declarati…
- `error` (object): Present when the call did not complete as asked. not_found: no published record answers. ambiguous: several records match and none was chosen (see candidates). invalid_input: an argument could not be…
- `evidence` (object): Present on a published profile. What this answer is built from, what it leaves out, and where to get the rest. The summary is a summary: the typed facts with their provenance are in the record at ful…
- `identity` (object): trooth_public_trust_profile, when the record serves it (contract member identity): legal_name, domain, registry_identifier, canonical_url and note.
- `next_cursor` (string|null): The list tools: send it back as cursor, with the same other arguments, for the next page; null on the last page.
- `outside_in` (object): trooth_outside_in_read only, with status observed: what this one read observed, field by field, the same observations as the summary. Each header and security.txt state is an observation of the respo…
- `page` (object): One page in a fixed, documented order. next_cursor is opaque: send it back as cursor with the same arguments, or null on the last page.
- `profile` (object): trooth_public_trust_profile with detail full, on a published record: the complete Trust Profile (format trooth.company-profile.v1), every section in the page's order or the sections asked for. Each s…
- `provenance` (string): Where this answer came from. An honest_absence is missing data, never a judgment about the subject. A read_failed means Trooth could not read its own record at this moment and asserts nothing either…
- `record_summary` (object): trooth_public_trust_profile, when the record serves it (contract member record_summary): record_version, last_witnessed_at, counts_by_provenance, sections_not_published, sections_on_request, checks_r…
- `results` (array): trooth_search_companies only: the published records that match, in the Network's own order. The `standing` field is Witnessed, Self-published or Unavailable (the witnessed feed did not answer; not a…
- `retry_after_seconds` (integer): status busy only: seconds to wait before trying trooth_outside_in_read again.
- `source` (object): What this answer was read from, which version of it, and the date that source itself carries. as_of is null when the source carries no date, never a guessed one.
- `status` (string): Machine-branchable outcome for this lookup.
- `subject` (string): The company, domain or token this answer is about.
- `summary` (string): The same answer as the text content, for display. Bounded; a longer answer is cut and says where the rest is.
- `token` (object): trooth_verify only: what the token and the ledger carry that Trooth did not write. key_id is the key id the token names; revoked_reason is the revoker's words; declared holds the issuing company's ow…
- `untrusted_text` (object): A10. present: this answer carries text Trooth did not write, in the fields content_trust lists (a company's own words, a provider's answer, bytes read from a website). flagged: the exact paths of tho…

### `trooth_verify` (~247 tokens)

Check a signed copy or a Trust Ledger Token

Check the two signatures on a Trust Ledger Token (tlt2. or tlt. form, or its JTI) against Trooth's own token ledger. A Trust Ledger Token is a signed token a company issues through Trooth about its own declared claims: the company signs the claims and Trooth signs the event of issuing it. Returns valid, invalid (a signature does not check out), expired or revoked, or unclaimed when no token matches; Trooth's signature covers the signing event, not the claim bytes and not the truth of the claims. With url in place of token, it checks a page Semanto signs against its signed copy: matches, differs or not signed, with the version, the signer and the proof address. It takes token or url, one of the two.

Input parameters:

- `hash` (string): With url: the sha256: digest of the markdown the caller was served, compared in place of a fetch.
- `token` (string): A Trust Ledger Token (tlt2. or tlt. form) or its JTI.
- `url` (string): A page Semanto signs (https), to check it against its signed copy.

### `trooth_semanto_status` (~39 tokens)

Setup status and next step

Setup status and next step. Sign in to Trooth, or send an agent key, to see what it takes and to use it.

### `trooth_semanto_setup_read_again` (~40 tokens)

Read the website again

Read the website again. Sign in to Trooth, or send an agent key, to see what it takes and to use it.

### `trooth_semanto_setup_choose` (~42 tokens)

Choose Basic or the full signed copy

Choose Basic or the full signed copy. Sign in to Trooth, or send an agent key, to see what it takes and to use it.

### `trooth_semanto_review_get` (~39 tokens)

The facts to review

The facts to review. Sign in to Trooth, or send an agent key, to see what it takes and to use it.

### `trooth_semanto_setup_confirm` (~39 tokens)

Confirm the reviewed version

Confirm the reviewed version. Sign in to Trooth, or send an agent key, to see what it takes and to use it.

### `trooth_semanto_sign` (~37 tokens)

Sign and publish

Sign and publish. Sign in to Trooth, or send an agent key, to see what it takes and to use it.

### `trooth_semanto_install_guide` (~43 tokens)

How to put it on the site

How to put it on the site. Sign in to Trooth, or send an agent key, to see what it takes and to use it.

### `trooth_semanto_install_check` (~38 tokens)

Check my site

Check my site. Sign in to Trooth, or send an agent key, to see what it takes and to use it.

### `trooth_semanto_page` (~63 tokens)

Read one signed page

One page of a company's signed copy as markdown, by its path on the company's site. No key needed.

Input parameters:

- `path` (string, required): The page's path, for example /pricing.
- `trooth_id` (string, required): The company's Trooth ID.

### `trooth_semanto_facts` (~98 tokens)

A company's signed facts

The company's signed facts as JSON-LD. For a large record, send page_size for one page of facts at a time and next_cursor back as cursor. No key needed.

Input parameters:

- `cursor` (string): Optional: the next_cursor of the page before.
- `page_size` (integer): Optional: 1 to 1,000 facts a page.
- `trooth_id` (string, required): The company's Trooth ID.

### `trooth_semanto_answer` (~185 tokens)

Answer a question from a company's signed record

One answer from the company's signed record: the signed sentence word for word, its value in standard form, its conditions, when it was signed, who stated it, the check link, the fact ID and the version hash. The question is matched in a fixed order (fact ID, the record's question map, words that name a fact type, word overlap) and matched_by names the step. More than one fitting fact, or a name two companies share, returns the candidates and picks none; nothing fitting returns not_published, never a nearby fact. No generated text. No key needed.

Input parameters:

- `company` (string, required): The company: its Trooth ID, its domain or its name.
- `fact_id` (string): Optional: a fact ID instead of a question.
- `question` (string): Optional: the question, at most 300 characters.

### `trooth_semanto_statement_check` (~191 tokens)

Check a statement against the signed record

Checks a statement about a company, sentence by sentence, against its current signed record by fixed rules with no model: supported (with the fact ID), differs (with the signed sentence), not_in_record, or not_checked for a sentence with no value a rule reads. It is meant for a price, an hour, a phone number or a date before it is repeated. To check one value before stating it, send fact_id and value instead: the answer is matches or differs by rule alone. No key needed.

Input parameters:

- `fact_id` (string): Check by value: the fact's ID.
- `statement` (string): The text to check, up to 4,000 characters.
- `trooth_id` (string, required): The company's Trooth ID.
- `value` (string): Check by value: the value as it will be stated, up to 1,000 characters.

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/troothllc-trooth-network/mcp#diagnostics

## Score history

- 2026-11-21: 68
- 2026-11-20: 68
- 2026-11-19: 68
- 2026-11-18: 68
- 2026-11-17: 68
- 2026-11-16: 68
- 2026-11-15: 68
- 2026-11-14: 68
- 2026-11-13: 69
- 2026-11-12: 69
- 2026-11-11: 68
- 2026-11-10: 69

## Common questions

### What is the Trooth MCP server?

Trooth is an MCP server listed in the public MCP registry as io.github.troothllc/trooth-network. Read any company's record on the Trooth Network, each fact labeled with who said it and when. This page covers its hosted endpoint (https://mcp.trooth.co).

### Is the Trooth MCP server safe to use?

Trooth scores 68 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Trooth MCP server expose?

Trooth exposes 42 tools: trooth_setup, trooth_hub_sync, enable_toolset, search_tools, call_tool, and 37 more. Their descriptions and schemas cost roughly 6,120 tokens of context every time the server is loaded.

### Does the Trooth MCP server require authentication?

No. We connected to Trooth without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the Trooth MCP server still maintained?

Trooth is marked deprecated in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the Trooth MCP server?

VerifyMCP ran 14 checks on Trooth, covering TLS certificate, Authorization, HTTPS, HSTS header and 10 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://mcp.trooth.co/
- Repository: https://github.com/troothllc/trooth-mcp
- Website: https://trooth.co/
- Changelog RSS feed: https://verifymcp.io/servers/troothllc-trooth-network/mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/troothllc-trooth-network/mcp/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/troothllc-trooth-network/mcp.json
- HTML version of this page: https://verifymcp.io/servers/troothllc-trooth-network/mcp
