# Bosco Labs: official records for AI agents (remote · api.boscolabs.dev)

Sanctions, KYB company checks, SEC EDGAR, EU law and more from official sources, each row sourced

- Trust score: 71/100 (medium)
- Change this week: +2
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- remote · `api.boscolabs.dev`: 71/100 (this document), [markdown](https://verifymcp.io/servers/bosco-labs-bosco-labs-data/api.md), [page](https://verifymcp.io/servers/bosco-labs-bosco-labs-data/api)
- remote · `api.boscolabs.dev`: 77/100, [markdown](https://verifymcp.io/servers/bosco-labs-bosco-labs-data/mcp-free.md), [page](https://verifymcp.io/servers/bosco-labs-bosco-labs-data/mcp-free)
- pypi · `bosco-labs-data-mcp`: 77/100, [markdown](https://verifymcp.io/servers/bosco-labs-bosco-labs-data/bosco-labs-data-mcp.md), [page](https://verifymcp.io/servers/bosco-labs-bosco-labs-data/bosco-labs-data-mcp)

## Channel facts

- Endpoint: `https://api.boscolabs.dev/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `0.4.0`

## 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**: 80/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - No authorisation is required to call this server. Nothing on it is marked destructive, but we can't show that an anonymous caller can only read: full credit needs the server to list at least one tool and every tool to be verifiably read-only. We use a best-effort check of tool names and descriptions to tell read-only tools from tools that change things, so some genuinely read-only tools may not get full credit. Names that start with a read verb (get, list, read, search and similar) and accurate readOnlyHint annotations help. Meet that to earn full credit, or require authorisation.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 59/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 2040 tokens (~340/item across 6 items; 6 tools + 0 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 14 days we've observed: 27 tool removals, 0 breaking changes, 0 auth/transport breaks, 6 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 (50% 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.
  - We read all 6 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 7 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).

## Install

### How do I install the Bosco Labs: official records for AI agents MCP server?

Bosco Labs: official records for AI agents is a hosted endpoint at https://api.boscolabs.dev/mcp, 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 bosco-labs-bosco-labs-data 'https://api.boscolabs.dev/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "bosco-labs-bosco-labs-data": {
      "url": "https://api.boscolabs.dev/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "bosco-labs-bosco-labs-data": {
      "type": "http",
      "url": "https://api.boscolabs.dev/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.bosco-labs-bosco-labs-data]
url = "https://api.boscolabs.dev/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "bosco-labs-bosco-labs-data": {
      "type": "remote",
      "url": "https://api.boscolabs.dev/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add bosco-labs-bosco-labs-data --url 'https://api.boscolabs.dev/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  bosco-labs-bosco-labs-data:
    url: "https://api.boscolabs.dev/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "bosco-labs-bosco-labs-data": {
      "Transport": "http",
      "Url": "https://api.boscolabs.dev/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add bosco-labs-bosco-labs-data -t streamable-http -u 'https://api.boscolabs.dev/mcp'
```

### Other

```json
{
  "mcpServers": {
    "bosco-labs-bosco-labs-data": {
      "type": "http",
      "url": "https://api.boscolabs.dev/mcp"
    }
  }
}
```

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 71, +1)

- [security] Tool “check_counterparty” rewrote its description, which is the text the model reads
- [functional improvement] Schema quality: 379 → 340
- [functional improvement] Tool “check_counterparty” now declares an output schema
- [functional] First check of Tool coverage: 50
- [functional] Schema quality: good → excellent
- [functional] New tool “describe_records_tool”
- [functional] New tool “find_sec_company”
- [functional] New tool “get_lei_record”

### 2026-11-17 (score 70, 0)

- [security regression] A breaking change shipped without a version bump: still 0.1.0
- [security regression] Tool “counterparty_check” was removed
- [security regression] Tool “plan_official_records_query” was removed
- [security regression] Tool “run_official_records_plan” was removed
- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “counterparty_check” rewrote its description, which is the text the model reads
- [functional] New tool “run_records_plan”
- [functional] New tool “plan_records_query”
- [functional] New tool “check_counterparty”
- [cosmetic] “counterparty_check” reworded the description of “country”

### 2026-11-16 (score 70, +1)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “run_official_records_plan” rewrote its description, which is the text the model reads
- [security] Tool “plan_official_records_query” rewrote its description, which is the text the model reads
- [cosmetic] Tool “counterparty_check” changed its title: Counterparty Check API: KYB checklist from official registers → Counterparty Check API: KYB, sanctions screening, insolvency

### 2026-11-13 (score 69, 0)

- [security] The server rewrote its instructions, which are the text every model session reads

### 2026-11-12 (score 69, 0)

- [security] The server rewrote its instructions, which are the text every model session reads

### 2026-11-10 (score 69, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [cosmetic] “counterparty_check” reworded the description of “country”

### 2026-11-09 (score 69, 0)

- [security regression] Stability: 0.03 → fail
- [security regression] A breaking change shipped without a version bump: still 0.1.0
- [security regression] Tool “sec_ownership” was removed
- [security regression] Tool “us_federal_awards” was removed
- [security regression] Tool “us_federal_rules” was removed
- [security regression] Tool “us_house_trades” was removed
- [security regression] Tool “vat_iban_check” was removed
- [security regression] Tool “describe_official_records_tool” was removed
- [security regression] Tool “company_hiring_signals” was removed
- [security regression] Tool “drug_safety” was removed
- [security regression] Tool “company_accounts” was removed
- [security regression] Tool “clinical_trials” was removed
- [security regression] Tool “uk_new_companies” was removed
- [security regression] Tool “tariff_lookup” was removed
- [security regression] Tool “company_check” was removed
- [security regression] Tool “dutch_address_lookup” was removed
- [security regression] Tool “dutch_case_law” was removed
- [security regression] Tool “esef_annual_reports” was removed
- [security regression] Tool “eu_law_lookup” was removed
- [security regression] Tool “eu_rates_inflation” was removed
- [security regression] Tool “eu_tenders” was removed
- [security regression] Tool “identifier_resolver” was removed
- [security regression] Tool “national_law_lookup” was removed
- [security regression] Tool “package_vulns” was removed
- [security regression] Tool “product_recalls” was removed
- [security regression] Tool “sanctions_screening” was removed
- [security regression] Tool “sec_company_financials” was removed
- [security regression] Tool “sec_company_screen” was removed
- [security regression] Tool “sec_filings_feed” was removed
- [security regression] Tool “sec_insider_trades” was removed
- [security regression] Tool “dod_contract_awards” was removed
- [security] Tool “plan_official_records_query” rewrote its description, which is the text the model reads
- [security] Tool “run_official_records_plan” rewrote its description, which is the text the model reads
- [functional improvement] Schema quality: 379 → 321
- [functional] Destructive annotations: 0 → pass
- [functional] New tool “counterparty_check”

### 2026-11-08 (score 69, 0)

- [security] Tool “sanctions_screening” rewrote its description, which is the text the model reads
- [security] Tool “company_check” rewrote its description, which is the text the model reads
- [functional improvement] Stability: unverified → 0.03
- [functional] New tool “vat_iban_check”
- [functional] New tool “run_official_records_plan”
- [functional] New tool “plan_official_records_query”
- [functional] New tool “describe_official_records_tool”
- [cosmetic] “sanctions_screening” added an optional parameter “addresses”
- [cosmetic] “company_check” reworded the description of “include”
- [cosmetic] “sanctions_screening” made “names” optional

## MCP tools (6)

### `check_counterparty` (~531 tokens)

Counterparty Check API: KYB, sanctions screening, insolvency

Can I do business with this company? One call checks a supplier, customer or partner against official sources and returns a checklist: pass, fail, review or unverified per check, each with its source link and time. To screen a vendor list, pass up to 10 companies in `companies`: one checklist per company, same price each.

Inputs: `query` for one company, or `companies` for up to 10 (a name or number each, or an object with these same fields). `country` narrows a name search to one register. `vatNumber`, `iban` and `walletAddresses` are the details the counterparty gave you: each adds a check (VAT registered to this company, IBAN valid and in its country, wallets not sanctioned); leave out what you do not have.

Read-only, no side effects. Charged per row returned, at the price below; a call that fails or runs out of time (55 s) is free. If one source does not answer, the rest still comes back and `warnings` names the gap. Limit: 60 calls a minute per key (10 on a new free key); past it the call is refused (rate_limited, free): retry after about 10 seconds.

Price: $0.05 per `counterparty-check`.

Input parameters:

- `companies` (array): Check a list of companies in one run instead of `query`: up to 10 companies, each a name or number ("Tesco PLC") or an object with the fields above
- `country` (string): ISO 3166 alpha-2 country code to disambiguate, e.g. "DK", "GB", "NO", "FR", "FI", "EE", "IE", "CZ", "LV", "LT", "SK", "SE", "BE", "NL", "US"
- `iban` (string): The bank account the counterparty wants to be paid on, e.g. "DK5000400440116243"
- `query` (string): The counterparty to check: a name (e.g. "Novo Nordisk") or an identifier: LEI
- `vatNumber` (string): The VAT number the counterparty gave you, with its country prefix, e.g. "DK24256790"
- `walletAddresses` (array): Wallet addresses the counterparty wants to be paid on, e.g. ["0x7FF9cFad3877F21d41Da833E2F775dB0569eE3D9"]

Output parameters:

- `balance_usd` (number|null)
- `charged_usd` (number|null)
- `count` (number|null)
- `duration_seconds` (number|null)
- `message` (string|null)
- `rows` (array)
- `tool` (string|null)
- `warnings` (array|null)

### `plan_records_query` (~316 tokens)

Plan an official-records query

Free, no key needed. Turns a question in plain words into the exact typed Bosco Labs calls that answer it: tool, input, why, and the maximum price of each step, plus what is not covered. Pass the user's question verbatim AND the entities you extracted from it (company_names, tickers, identifiers, countries, dates, law references...): the planner trusts your entities over its own parsing and puts them in the right input fields. The plan echoes them in entities_used and lists in `missing` what to pass for a better plan. Rules and examples first; only when they are unsure and you passed no entities does a small AI step extract the entities (budget-capped, marked `assisted`). Same steps, same plan_id. Run it with run_records_plan (paid, max_usd) or call the listed tools directly. Low confidence returns the top 3 candidate tools instead. Questions are kept 30 days in a cut-down form; leave out personal details.

Input parameters:

- `entities` (object): What you (the calling agent) already extracted from the user's question. Extract these from the question yourself; the planner trusts them over its own parsing and puts them in the right input fields…
- `question` (string, required): The user's question, verbatim, e.g. "Is Rosneft on the EU sanctions list?" (English, Dutch, German or French).

### `run_records_plan` (~208 tokens)

Run an official-records plan

Runs a plan from plan_records_query (its plan_id) or explicit steps, all steps in parallel, under a hard max_usd: if the worst case could exceed it, nothing runs. Each step is charged to the caller's key exactly like a direct tool call (failed steps are free). Returns one section per step (via_tool, via_input, rows with source_url), a receipt and the citations. Reaches all 30 tools; to list them as MCP tools, add ?tools=all to the MCP URL.

Input parameters:

- `max_usd` (number, required): Hard cap in USD: refused, nothing run, if the steps' worst case could cost more.
- `plan_id` (string): The plan_id plan_records_query returned (or pass steps).
- `steps` (array): Explicit steps instead of a plan_id: [{tool, input}], tool a name from GET /v1/tools (eu-tenders or search_eu_tenders), input that tool's input.

### `describe_records_tool` (~106 tokens)

Describe an official-records tool

Free, no key needed. One Bosco Labs tool's full input schema, example input, example rows, output fields, price per row, caps and sources. Use it before calling a tool whose input you are unsure of: the tool list shortens each input field to one sentence, and this shows every field, its allowed values and a working example. An unknown name returns the closest names.

Input parameters:

- `name` (string, required): Tool name, e.g. search_eu_tenders.

### `get_lei_record` (~355 tokens)

LEI lookup (GLEIF, free)

Free, no API token needed. Legal Entity Identifier (LEI) lookup in the official GLEIF register: about 2.9 million companies, funds and other legal entities worldwide (source: GLEIF API, CC0 open data). Use it to identify a company: its exact legal name, LEI, whether it is active, its jurisdiction and country, its local registration number and its ultimate parent company.

Pass an LEI (exact record) or a company name (best matches first; check `legal_name`).

Returns one row per entity: lei, legal_name, entity_status (ACTIVE/INACTIVE), registration_status (ISSUED, LAPSED, ...), jurisdiction, legal_address_country, legal_address_city, headquarters_country, registered_as (local registry number), ultimate_parent (lei, legal_name; when reported), source_url (the GLEIF record; cite it) and next_steps (the paid Bosco Labs call for this company's financials: the tool and arguments to call next).

Inputs: `query` is either a 20-character LEI (one exact record) or a company name (best matches first); `max_results` (1-5, default 3) applies to a name search only. Read-only, no key, no charge; at most 30 free lookups a minute per caller. Nothing found: an error that says what to try.

Input parameters:

- `max_results` (integer): Max companies for a name search (1-5).
- `query` (string, required): A 20-character LEI, e.g. 724500Y6DUVHQD6OXN27, or a company name, e.g. "Heineken"

Output parameters:

- `balance_usd` (number|null)
- `charged_usd` (number|null)
- `count` (number|null)
- `duration_seconds` (number|null)
- `message` (string|null)
- `rows` (array)
- `tool` (string|null)
- `warnings` (array|null)

### `find_sec_company` (~308 tokens)

SEC company lookup: ticker to CIK (free)

Free, no API token needed. Identify a US-listed (SEC-registered) company: CIK, official name, ticker(s), exchange(s), industry (SIC code) and state of incorporation (source: SEC EDGAR company tickers file and submissions API). Use it to turn a ticker or name into the CIK that SEC tools take, or to check which exchange a stock trades on.

Pass a ticker, a CIK or part of the name (best matches first, then by company size; check `name`). A name search covers companies with a listed ticker (about 10,000); a CIK also finds filers without one.

Returns one row per company: cik, name, tickers, exchanges, sic, sic_description, state_of_incorporation, fiscal_year_end (MMDD), source_url (the EDGAR company page; cite it) and next_steps (the paid Bosco Labs calls for its filings and insider trades).

Inputs: `query` is a ticker, a CIK or part of the name; `max_results` (1-5, default 3) applies to a name search only. Read-only, no key, no charge; at most 30 free lookups a minute per caller.

Input parameters:

- `max_results` (integer): Max companies for a name search (1-5).
- `query` (string, required): A ticker (AAPL), a CIK (320193) or part of the company name ("apple")

Output parameters:

- `balance_usd` (number|null)
- `charged_usd` (number|null)
- `count` (number|null)
- `duration_seconds` (number|null)
- `message` (string|null)
- `rows` (array)
- `tool` (string|null)
- `warnings` (array|null)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/bosco-labs-bosco-labs-data/api#diagnostics

## Score history

- 2026-11-21: 71
- 2026-11-20: 71
- 2026-11-19: 71
- 2026-11-18: 71
- 2026-11-17: 70
- 2026-11-16: 70
- 2026-11-15: 69
- 2026-11-14: 69
- 2026-11-13: 69
- 2026-11-12: 69
- 2026-11-11: 69
- 2026-11-10: 69
- 2026-11-09: 69
- 2026-11-08: 69
- 2026-11-07: 69

## Common questions

### What is the Bosco Labs: official records for AI agents MCP server?

Bosco Labs: official records for AI agents is an MCP server listed in the public MCP registry as io.github.Bosco-Labs/bosco-labs-data. Sanctions, KYB company checks, SEC EDGAR, EU law and more from official sources, each row sourced. This page covers its hosted endpoint (https://api.boscolabs.dev/mcp).

### Is the Bosco Labs: official records for AI agents MCP server safe to use?

Bosco Labs: official records for AI agents scores 71 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 Bosco Labs: official records for AI agents MCP server expose?

Bosco Labs: official records for AI agents exposes 6 tools: check_counterparty, plan_records_query, run_records_plan, describe_records_tool, get_lei_record, find_sec_company. Their descriptions and schemas cost roughly 1,824 tokens of context every time the server is loaded.

### Does the Bosco Labs: official records for AI agents MCP server require authentication?

No. We connected to Bosco Labs: official records for AI agents without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the Bosco Labs: official records for AI agents MCP server still maintained?

Bosco Labs: official records for AI agents is still listed as active 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 Bosco Labs: official records for AI agents MCP server?

VerifyMCP ran 13 checks on Bosco Labs: official records for AI agents, covering TLS certificate, Authorization, HTTPS, HSTS header and 9 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://api.boscolabs.dev/mcp
- Authorisation metadata: https://api.boscolabs.dev/.well-known/oauth-protected-resource/mcp
- Repository: https://github.com/Bosco-Labs/bosco-labs-data
- Website: https://boscolabs.dev/
- Changelog RSS feed: https://verifymcp.io/servers/bosco-labs-bosco-labs-data/api.xml
- Changelog JSON feed: https://verifymcp.io/servers/bosco-labs-bosco-labs-data/api/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/bosco-labs-bosco-labs-data/api.json
- HTML version of this page: https://verifymcp.io/servers/bosco-labs-bosco-labs-data/api
