# Transept (remote · app.transept.ai)

Translate documents with glossaries, style guides and translation memory; run and read back jobs.

- Trust score: 73/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- remote · `app.transept.ai`: 73/100 (this document), [markdown](https://verifymcp.io/servers/ai-transept-transept/api-public-v1-mcp.md), [page](https://verifymcp.io/servers/ai-transept-transept/api-public-v1-mcp)

## Channel facts

- Endpoint: `https://app.transept.ai/api/public/v1/mcp`
- Transports: `streamable-http`
- Auth: `required`
- Version: `1.0.0`

## 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**: 94/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token.
  - 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.
  - The authorisation server offers only Dynamic Client Registration (RFC 7591), which MCP 2026-07-28 deprecated in favour of Client ID Metadata Documents.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 56/100
  - AI-judged instruction clarity (fair).
  - Tool/resource definitions use about 2587 tokens (~61/item across 42 items; 42 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 13/100
  - Stability observed for 4 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 67/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 0% of tool parameters carry a description.
- **Tool Safety**: 75/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - 0 of 2 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "document_delete" implies "delete" and declares no destructiveHint at all, which the MCP spec reads as destructive by default.
  - An AI judge read all 43 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

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

Transept is a hosted endpoint at https://app.transept.ai/api/public/v1/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 ai-transept-transept 'https://app.transept.ai/api/public/v1/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "ai-transept-transept": {
      "url": "https://app.transept.ai/api/public/v1/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "ai-transept-transept": {
      "type": "http",
      "url": "https://app.transept.ai/api/public/v1/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.ai-transept-transept]
url = "https://app.transept.ai/api/public/v1/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "ai-transept-transept": {
      "type": "remote",
      "url": "https://app.transept.ai/api/public/v1/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add ai-transept-transept --url 'https://app.transept.ai/api/public/v1/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  ai-transept-transept:
    url: "https://app.transept.ai/api/public/v1/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "ai-transept-transept": {
      "Transport": "http",
      "Url": "https://app.transept.ai/api/public/v1/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add ai-transept-transept -t streamable-http -u 'https://app.transept.ai/api/public/v1/mcp'
```

### Other

```json
{
  "mcpServers": {
    "ai-transept-transept": {
      "type": "http",
      "url": "https://app.transept.ai/api/public/v1/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.

### 2026-11-19 (score 73, +1)

No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-18 (score 72, 0)

- [functional improvement] Stability: unverified → 0.03

### 2026-11-17 (score 72, +34)

- [security improvement] Transport: unverified → pass
- [security improvement] Injection markers: unverified → pass
- [security] First check of Judged manipulation: pass
- [security] Authorization: Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token.
- [functional improvement] MCP protocol: unverified → pass
- [functional improvement] Tool coverage: unverified → 100
- [functional] First check of Tool coverage: 0
- [functional] First check of Schema quality: fail
- [functional] First check of Schema quality: fair
- [functional] First check of Destructive annotations: 0
- [functional] First check of Schema quality: pass

### 2026-11-16 (score 38)

First indexed and scored.

## MCP tools (42)

### `me_get` (~32 tokens)

The account this API key belongs to, the key in use, its scopes, and the credit balance available to spend.

### `project_list` (~16 tokens)

List projects you can file documents into.

### `project_create` (~69 tokens)

Create a project (pass team_id for a team project; parent_id for a sub-project of a top-level project — same owner, one level deep).

Input parameters:

- `description` (string)
- `name` (string, required)
- `parent_id` (string)
- `team_id` (string)

### `document_list` (~55 tokens)

List documents, newest first. One row per language version; siblings share language_group_id.

Input parameters:

- `cursor` (string)
- `limit` (integer)
- `project_id` (string)
- `search` (string)

### `document_get` (~23 tokens)

Document metadata + its language group.

Input parameters:

- `document_id` (string, required)

### `document_create` (~95 tokens)

Create a document from inline text/markdown/html and fan out target languages.

Input parameters:

- `content` (string, required)
- `content_type` (string)
- `glossary_ids` (array)
- `project_id` (string)
- `source_language` (string)
- `styleguide_version_ids` (array)
- `target_language` (string)
- `target_languages` (array)
- `title` (string)

### `document_update` (~50 tokens)

Update document metadata (title / category / project filing).

Input parameters:

- `document_category` (string)
- `document_id` (string, required)
- `project_id` (string)
- `title` (string)

### `document_delete` (~24 tokens)

Delete ONE language version of a document.

Input parameters:

- `document_id` (string, required)

### `document_add_languages` (~52 tokens)

Add target-language sibling versions to a document (grows its language group). Idempotent — languages already in the group are skipped.

Input parameters:

- `document_id` (string, required)
- `target_languages` (array, required)

### `document_content_get` (~44 tokens)

Read content as json | markdown | html | strings (binary formats: use the REST content endpoint).

Input parameters:

- `document_id` (string, required)
- `format` (string)

### `document_attach_resources` (~53 tokens)

Attach/detach glossaries + style-guide versions ([] = detach all of that kind).

Input parameters:

- `document_id` (string, required)
- `glossary_ids` (array)
- `styleguide_version_ids` (array)

### `document_resync_source` (~75 tokens)

Re-parse the document from its STORED source file and re-align blocks (unchanged blocks keep translations). The file owns sources: block_update edits are NOT in the file, so a resync overwrites them — resync when the source file changed, not after block edits.

Input parameters:

- `document_id` (string, required)

### `strings_import` (~142 tokens)

Import a Transept Strings corpus (.tstrings.json shape). Locales: explicit params win; with none named the corpus header target_locales apply (no locales anywhere → sync no_target_languages). A single-locale import is ONE version row carrying the corpus (language_group_id null until more locales join). Async: returns processing_job_id — poll processing_job_get. Translations the corpus carries land active and bill nothing.

Input parameters:

- `document` (object, required)
- `document_category` (string)
- `project_id` (string)
- `target_language` (string)
- `target_languages` (array)
- `use_target_locales` (boolean)

### `strings_reimport` (~96 tokens)

Re-submit the WHOLE current corpus against an existing master: the server diffs by unit id and answers with {units, versions[], delta_block_ids} — pass delta_block_ids VERBATIM as group_run_start.block_ids to run only the delta. Unchanged units keep their translations. Synchronous, bills nothing.

Input parameters:

- `document` (object, required)
- `document_id` (string, required)
- `overwrite_targets` (boolean)

### `block_update` (~73 tokens)

Edit a block's SOURCE content (Slate JSON). Note: edits live on the block, not in the stored source file — a later document_resync_source re-parses the file and overwrites them.

Input parameters:

- `block_id` (string, required)
- `block_type` (string)
- `original_content` (array, required)

### `block_translation_update` (~39 tokens)

Edit a translation (Slate JSON); it stays active and feeds TM.

Input parameters:

- `translated_content` (array, required)
- `translation_id` (string, required)

### `processing_job_get` (~51 tokens)

Poll an import job — the id returned by strings_import (or a REST file upload / Figma import). Runs are NOT processing jobs: poll those with run_get.

Input parameters:

- `job_id` (string, required)

### `workflow_list` (~19 tokens)

Saved workflows (ids go into run_start.workflow_id).

### `workflow_template_list` (~21 tokens)

Built-in templates (ids go into run_start.template_id).

### `run_estimate` (~72 tokens)

Price a run (free, no side effects). Costs are in credits.

Input parameters:

- `block_ids` (array)
- `document_id` (string, required)
- `parallel_mode` (boolean)
- `template_id` (string)
- `updated_only` (boolean)
- `workflow_id` (string)

### `run_start` (~86 tokens)

Start a workflow run. BILLS CREDITS — without confirm:true this only returns the estimate and starts nothing.

Input parameters:

- `block_ids` (array)
- `confirm` (boolean)
- `document_id` (string, required)
- `parallel_mode` (boolean)
- `template_id` (string)
- `updated_only` (boolean)
- `workflow_id` (string)

### `run_get` (~36 tokens)

Poll a run: status, progress, and (when gate_pending) the review-gate digest.

Input parameters:

- `job_id` (string, required)

### `run_list` (~58 tokens)

List runs, newest first. Paged like document_list: {data, has_more, next_cursor}.

Input parameters:

- `cursor` (string)
- `document_id` (string)
- `limit` (integer)
- `status` (string)

### `run_cancel` (~42 tokens)

Cancel a run (idempotent — answers with the resulting status, so an already-terminal run reads as what it already was).

Input parameters:

- `job_id` (string, required)

### `run_gate_approve` (~126 tokens)

Approve a review gate. Continuing downstream steps BILLS CREDITS — without confirm:true this only returns what continuing will cost (up to max_credits) and continues nothing. Omit block_ids to continue over ALL blocks the run processed; pass a subset to continue only those; pass [] to close the gate WITHOUT running (or billing) anything further (no confirm needed). No downstream steps → error no_continuation (use run_gate_acknowledge).

Input parameters:

- `block_ids` (array)
- `confirm` (boolean)
- `job_id` (string, required)

### `run_gate_acknowledge` (~27 tokens)

Dismiss a review gate without continuing.

Input parameters:

- `job_id` (string, required)

### `group_run_estimate` (~80 tokens)

Price a multi-language fan-out (free).

Input parameters:

- `block_ids`
- `document_id` (string, required)
- `include_untranslated` (boolean)
- `parallel_mode` (boolean)
- `target_languages` (required)
- `template_id` (string)
- `updated_only` (boolean)
- `workflow_id` (string)

### `group_run_start` (~99 tokens)

Run a workflow across many languages. BILLS CREDITS — without confirm:true this only returns the estimate.

Input parameters:

- `block_ids`
- `confirm` (boolean)
- `document_id` (string, required)
- `include_untranslated` (boolean)
- `parallel_mode` (boolean)
- `target_languages` (required)
- `template_id` (string)
- `updated_only` (boolean)
- `workflow_id` (string)

### `group_run_get` (~30 tokens)

Poll a group run: rollup + per-language lanes.

Input parameters:

- `group_run_id` (string, required)

### `group_run_list` (~30 tokens)

List group runs, newest first.

Input parameters:

- `cursor` (string)
- `limit` (integer)

### `group_run_cancel` (~28 tokens)

Cancel every still-active lane of a group run.

Input parameters:

- `group_run_id` (string, required)

### `glossary_list` (~19 tokens)

List glossaries (yours + your teams’).

### `glossary_get` (~22 tokens)

One glossary with its terms.

Input parameters:

- `glossary_id` (string, required)

### `glossary_create` (~55 tokens)

Create a glossary (optionally seed terms).

Input parameters:

- `description` (string)
- `name` (string)
- `primary_target_language` (string)
- `source_language` (string)
- `terms` (array)

### `glossary_terms_add` (~34 tokens)

Add terms in bulk; the answer is counts.

Input parameters:

- `glossary_id` (string, required)
- `terms` (array, required)

### `styleguide_list` (~22 tokens)

List style guides with their ACTIVE version id (what documents consume).

### `styleguide_get` (~28 tokens)

One style guide incl. the active version content.

Input parameters:

- `styleguide_id` (string, required)

### `tm_query` (~66 tokens)

Query translation memory (hybrid retrieval; free).

Input parameters:

- `cross_language` (boolean)
- `document_id` (string)
- `query_text` (string, required)
- `source_lang` (string, required)
- `target_lang` (string, required)
- `top_k` (integer)

### `tm_import` (~49 tokens)

Import a TMX/XLIFF file’s text into translation memory (free; async embeddings).

Input parameters:

- `content` (string, required)
- `format` (string, required)
- `project_id` (string)

### `webhook_list` (~16 tokens)

List outgoing webhook endpoints (no secrets).

### `webhook_create` (~43 tokens)

Register an outgoing webhook. The response carries the signing secret ONCE.

Input parameters:

- `description` (string)
- `events` (array)
- `url` (string, required)

### `webhook_delete` (~22 tokens)

Delete an outgoing webhook endpoint.

Input parameters:

- `webhook_id` (string, required)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/ai-transept-transept/api-public-v1-mcp#diagnostics

## Score history

- 2026-11-21: 73
- 2026-11-20: 73
- 2026-11-19: 73
- 2026-11-18: 72
- 2026-11-17: 72
- 2026-11-16: 38

## Common questions

### What is the Transept MCP server?

Transept is an MCP server listed in the public MCP registry as ai.transept/transept. Translate documents with glossaries, style guides and translation memory; run and read back jobs. This page covers its hosted endpoint (https://app.transept.ai/api/public/v1/mcp).

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

Transept scores 73 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 Transept MCP server expose?

Transept exposes 42 tools: me_get, project_list, project_create, document_list, document_get, and 37 more. Their descriptions and schemas cost roughly 2,119 tokens of context every time the server is loaded.

### Does the Transept MCP server require authentication?

Yes. Transept asked us for credentials when we connected, so you will need to authorise it in your MCP client before it can do anything.

### Is the Transept MCP server still maintained?

Transept 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 Transept MCP server?

VerifyMCP ran 13 checks on Transept, 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://app.transept.ai/api/public/v1/mcp
- Website: https://transept.ai/help/connect-ai-assistant
- Changelog RSS feed: https://verifymcp.io/servers/ai-transept-transept/api-public-v1-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/ai-transept-transept/api-public-v1-mcp/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/ai-transept-transept/api-public-v1-mcp.json
- HTML version of this page: https://verifymcp.io/servers/ai-transept-transept/api-public-v1-mcp
