# Legesher (pypi · legesher-mcp)

Translate Python between English keywords and your language, with round-trip verification.

- Trust score: 66/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- pypi · `legesher-mcp`: 66/100 (this document), [markdown](https://verifymcp.io/servers/legesher-legesher-mcp/legesher-mcp.md), [page](https://verifymcp.io/servers/legesher-legesher-mcp/legesher-mcp)

## Channel facts

- Registry: `pypi`
- Package: `legesher-mcp`
- Version: `0.3.0`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, 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 2026-11-05.

- **Supply Chain Security**: 50/100
  - Malware scan not yet available for this package.
  - No known CVEs affecting this package version or its production dependencies.
  - Runs pdm.backend at install time, a recognised build step with no custom scripting around it.
  - 3 of 49 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 100/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to legesher/legesher).
  - Clear OSI-approved license (Apache-2.0).
  - Actively maintained (last published 1 days ago).
  - Publishes a security disclosure policy (SECURITY.md).
- **Schema Quality & AI Usability**: 73/100
  - 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 1334 tokens (~190/item across 7 items; 6 tools + 1 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability not yet verified: not enough scan history yet (needs a 30-day window).
- **Tool Coverage**: 93/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 77% of tool parameters carry a description.
  - Structured output schemas are declared (100% 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 8 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.

**Unverified: 1 category.** A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.

## Install

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

Legesher runs locally as a PyPI package, launched with uvx legesher-mcp. 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 legesher-legesher-mcp -- uvx legesher-mcp
```

### Cursor

```json
{
  "mcpServers": {
    "legesher-legesher-mcp": {
      "command": "uvx",
      "args": [
        "legesher-mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "legesher-legesher-mcp": {
      "command": "uvx",
      "args": [
        "legesher-mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add legesher-legesher-mcp -- uvx legesher-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "legesher-legesher-mcp": {
      "type": "local",
      "command": [
        "uvx",
        "legesher-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add legesher-legesher-mcp --command uvx --arg legesher-mcp
```

### Hermes

```yaml
mcp_servers:
  legesher-legesher-mcp:
    command: "uvx"
    args: ["legesher-mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "legesher-legesher-mcp": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "legesher-mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add legesher-legesher-mcp -t stdio -c uvx -a legesher-mcp
```

### Other

```json
{
  "mcpServers": {
    "legesher-legesher-mcp": {
      "command": "uvx",
      "args": [
        "legesher-mcp"
      ]
    }
  }
}
```

## 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.

### 2026-11-03 (score 66)

First indexed and scored.

## MCP tools (6)

### `translate_to_native` (~231 tokens)

Translate Python source code into the caller's natural language.

Accepts complete files and fragments alike — there is no separate
snippet mode to choose. When neither language is English the engine
pivots through English (packs are authored against English) and the
result reports `pivoted: true`. Formatting, comments, identifiers, and
string contents are preserved; only keywords, builtins, exceptions,
and reserved words are translated.

Input parameters:

- `python_version` (string)
- `source` (string, required): Python source text (a complete file or a fragment).
- `source_lang` (string): ISO 639 language code from the legesher://languages catalog.
- `target_lang` (string, required): ISO 639 language code from the legesher://languages catalog.
- `verify_round_trip` (boolean): Also verify the round trip: translate the result back, compare byte-for-byte, and check the canonical form compiles. Use this instead of comparing two renderings yourself — exact comparison over a lo…

Output parameters:

- `pivoted` (boolean): True when translated via the English pivot (X→en→Y): the engine has no direct X→Y path, so both hops rode language packs and round-trip fidelity depends on both.
- `python_version` (string)
- `round_trip`: Round-trip verification result. Present only when the tool was called with verify_round_trip: true; absent otherwise.
- `source_lang` (string)
- `target_lang` (string)
- `translated_source` (string)

### `translate_to_source` (~196 tokens)

Translate native-language code back to English-keyword Python.

The inverse of translate_to_native and the round-trip half of the
collaboration story: the output is the form the interpreter, the
toolchain, and upstream collaborators use. The target is fixed to
English — for a general X→Y translation call translate_to_native with
an explicit source_lang.

Input parameters:

- `python_version` (string)
- `source` (string, required): Python source text (a complete file or a fragment).
- `source_lang` (string, required): ISO 639 language code from the legesher://languages catalog.
- `verify_round_trip` (boolean): Also verify the round trip: translate the result back, compare byte-for-byte, and check the canonical form compiles. Use this instead of comparing two renderings yourself — exact comparison over a lo…

Output parameters:

- `pivoted` (boolean): True when translated via the English pivot (X→en→Y): the engine has no direct X→Y path, so both hops rode language packs and round-trip fidelity depends on both.
- `python_version` (string)
- `round_trip`: Round-trip verification result. Present only when the tool was called with verify_round_trip: true; absent otherwise.
- `source_lang` (string)
- `target_lang` (string)
- `translated_source` (string)

### `explain_in_language` (~165 tokens)

Return the engine facts needed to explain code in a language.

Deterministic material only — construct-by-construct keyword mappings
for the constructs actually present in the code, and a static
(compile-only, never executed) validation result. YOU compose the
explanation from these facts; the tool never generates prose, so
engine truth stays distinct from narration.

Input parameters:

- `code_lang` (string): ISO 639 language code from the legesher://languages catalog.
- `explain_in` (string, required): ISO 639 language code from the legesher://languages catalog.
- `include_translation` (boolean): Also return the source translated into explain_in.
- `python_version` (string)
- `source` (string, required): Python source text (a complete file or a fragment).

Output parameters:

- `code_lang` (string)
- `constructs` (array)
- `explain_in` (string)
- `python_version` (string)
- `translated_source`: The code translated into explain_in, when requested.
- `validation`

### `lookup_keyword_mapping` (~116 tokens)

Look up how specific terms render in a language.

The quick-reference tool: pass English terms for a filtered answer, or
omit them for the language's full mapping. For browsing the whole
dictionary prefer the legesher://languages/{lang}/terms resource.

Input parameters:

- `language` (string, required): ISO 639 language code from the legesher://languages catalog.
- `python_version` (string)
- `terms`: English terms to look up (e.g. ['def', 'print']). Omit for the full mapping.

Output parameters:

- `language` (string)
- `mappings`
- `python_version` (string)
- `unmatched_terms` (array): Requested terms with no entry in any category.

### `validate_identifier_translations` (~262 tokens)

Check proposed identifier translations before you record them.

Packs make keywords deterministic; identifiers have no pack, so this is
the guarantee they get instead. Each proposal is checked for a name the
pack has reserved, for drift (one concept acquiring a second term) and
for collision (one term acquiring a second concept) — against both the
existing glossary and the rest of the batch.

Returns a verdict per proposal plus ``accepted``, the conflict-free
subset as a mapping ready to merge into the project glossary. Recording
it is the caller's to do: this server writes nothing.

Input parameters:

- `glossary`: The project's existing identifier translations for this one language, flat: {source_identifier: translation}. If your repo has .legesher/identifier_map.json, that file is nested by language — send th…
- `proposals` (array, required): Proposed identifier translations to check, as source/translation pairs.
- `python_version` (string)
- `target_lang` (string, required): ISO 639 language code from the legesher://languages catalog.

Output parameters:

- `accepted` (object): The conflict-free proposals as a source -> translation mapping, ready to merge into the project glossary. The caller persists it; this server writes nothing. Both sides are NFC-normalized per the sto…
- `python_version` (string)
- `results` (array)
- `target_lang` (string)

### `map_existing_codebase` (~183 tokens)

Report what translating a codebase into a language would look like.

The legibility report: per-file translatability, construct counts,
constructs the target language's pack cannot map yet (they stay
English after translation), and an aggregate coverage summary.
Orchestrates per-file translate + validate; for a large repo, iterate
in batches of files across calls.

Input parameters:

- `files` (array, required): Files to map. Paths are report labels only — they are content, never logged, never used to touch the filesystem.
- `include_translations` (boolean): Also return the translated files (default: report only).
- `python_version` (string)
- `source_lang` (string): ISO 639 language code from the legesher://languages catalog.
- `target_lang` (string, required): ISO 639 language code from the legesher://languages catalog.

Output parameters:

- `files` (array)
- `python_version` (string)
- `source_lang` (string)
- `summary`
- `target_lang` (string)
- `translated_files`

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Dependencies. The full working is on the page: https://verifymcp.io/servers/legesher-legesher-mcp/legesher-mcp#diagnostics

## Score history

- 2026-11-05: 66
- 2026-11-04: 66
- 2026-11-03: 66

## Common questions

### What is the Legesher MCP server?

Legesher is an MCP server listed in the public MCP registry as io.github.legesher/legesher-mcp. Translate Python between English keywords and your language, with round-trip verification. This page covers its PyPI package (legesher-mcp).

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

Legesher scores 66 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 5 November 2026. Its build provenance is signed and verified. 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 Legesher MCP server expose?

Legesher exposes 6 tools: translate_to_native, translate_to_source, explain_in_language, lookup_keyword_mapping, validate_identifier_translations, map_existing_codebase. Their descriptions and schemas cost roughly 1,153 tokens of context every time the server is loaded.

### Is the Legesher MCP server still maintained?

Legesher is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the Legesher MCP server under?

Legesher declares the Apache-2.0 licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- PyPI project: https://pypi.org/project/legesher-mcp/
- Socket report: https://socket.dev/pypi/package/legesher-mcp
- Repository: https://github.com/legesher/legesher
- Website: https://legesher.io/
- Changelog RSS feed: https://verifymcp.io/servers/legesher-legesher-mcp/legesher-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/legesher-legesher-mcp/legesher-mcp.json
- HTML version of this page: https://verifymcp.io/servers/legesher-legesher-mcp/legesher-mcp