# io.github.x51xxx/codex-mcp-tool (npm · @trishchuk/codex-mcp-tool)

MCP server bridging AI assistants to OpenAI Codex CLI for code analysis and review

- Trust score: 66/100 (medium)
- Change this week: +16
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-08-03

## Components

- npm · `@trishchuk/codex-mcp-tool`: 66/100 (this document), [markdown](https://verifymcp.io/servers/x51xxx-codex-mcp-tool/trishchuk-codex-mcp-tool.md), [page](https://verifymcp.io/servers/x51xxx-codex-mcp-tool/trishchuk-codex-mcp-tool)

## Channel facts

- Registry: `npm`
- Package: `@trishchuk/codex-mcp-tool`
- Version: `2.2.0`
- 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 2026-08-03.

- **Supply Chain Security**: 87/100
  - No malware found by supply-chain analysis.
  - Only part of the dependency tree could be resolved (95 of 99), so this covers what we could see, not the whole tree.
  - No install/post-install scripts declared.
  - Only part of the dependency tree could be resolved (95 of 99), so this covers what we could see, not the whole tree.
- **Provenance & Transparency**: 45/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 20 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 66/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (fair).
  - Context-footprint check failed: tool/resource definitions use about 2333 tokens (~179/item across 13 items; 13 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 23/100
  - Stability observed for 7 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 95/100
  - 92% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 99% of tool parameters carry a description.
  - Structured output schemas are declared (46% of tools); any adoption earns full credit.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### Claude

```bash
claude mcp add x51xxx-codex-mcp-tool -- npx -y @trishchuk/codex-mcp-tool
```

### Codex

```bash
codex mcp add x51xxx-codex-mcp-tool -- npx -y @trishchuk/codex-mcp-tool
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "x51xxx-codex-mcp-tool": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "@trishchuk/codex-mcp-tool"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add x51xxx-codex-mcp-tool --command npx --arg -y --arg @trishchuk/codex-mcp-tool
```

### Hermes

```yaml
mcp_servers:
  x51xxx-codex-mcp-tool:
    command: "npx"
    args: ["-y", "@trishchuk/codex-mcp-tool"]
```

### Other

```json
{
  "mcpServers": {
    "x51xxx-codex-mcp-tool": {
      "command": "npx",
      "args": [
        "-y",
        "@trishchuk/codex-mcp-tool"
      ]
    }
  }
}
```

## 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.

### 2026-08-03 (score 66, +3)

- [functional improvement] Stability: unverified → 0.23

### 2026-08-02 (score 63, +58)

- [security regression] Provenance: unverified → fail
- [security improvement] Known CVEs: unverified → partial
- [security improvement] Install scripts: unverified → pass
- [security improvement] Malware scan: unverified → pass
- [security] Stability: Stability not yet verified: not enough scan history yet (needs a 30-day window).
- [functional regression] Security disclosure: fail → unverified
- [functional improvement] License: unverified → pass
- [functional improvement] Schema quality: unverified → fair
- [functional improvement] Tool coverage: unverified → 92
- [functional improvement] Schema quality: unverified → 100
- [functional improvement] Dependency health: unverified → partial
- [functional improvement] Maintenance: unverified → pass
- [functional improvement] MCP protocol: unverified → pass
- [functional] Licence: MIT

### 2026-08-01 (score 5, −19)

- [functional regression] Tool coverage: 92 → unverified
- [functional regression] Schema quality: 100 → unverified

### 2026-07-31 (score 24, −8)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-30 (score 32, −18)

- [security regression] Malware scan: pass → unverified

### 2026-07-27 (score 50)

First indexed and scored.

## MCP tools (13)

### `ask-codex` (~860 tokens)

Execute Codex CLI with file analysis (@syntax), skills ($syntax), model selection, and safety controls. Supports changeMode.

Input parameters:

- `addDirs` (array): Additional writable directories beyond workspace (e.g., ["/tmp", "/var/log"]). Useful for monorepos and multi-directory projects.
- `approval` (string): Approval policy: untrusted, on-failure, on-request, never
- `approvalPolicy` (string): Approval: never, on-request, on-failure, untrusted
- `cd` (string): Working directory
- `changeMode` (boolean): Return structured OLD/NEW edits for refactoring
- `chunkCacheKey` (string): Cache key for continuation
- `chunkIndex` (number): Chunk index (1-based)
- `config`: Configuration overrides as 'key=value' string or object
- `disableFeatures` (array): Disable feature flags (repeatable). Equivalent to -c features.<name>=false
- `enableFeatures` (array): Enable feature flags (repeatable). Equivalent to -c features.<name>=true
- `fullAuto` (boolean): Full automation mode
- `image`: Optional image file path(s) to include with the prompt
- `includeMetadata` (boolean): Include configuration metadata in response
- `includeThinking` (boolean): Include reasoning/thinking section in response
- `localProvider` (string): Specify which local provider to use (lmstudio or ollama). Automatically enables --oss if not set. If omitted with --oss, uses config default or shows selection.
- `model` (string): Model: gpt-5.4, gpt-5.3-codex, gpt-5.2-codex, gpt-5.1-codex-max, gpt-5.1-codex-mini, gpt-5.2. Default: gpt-5.4
- `oss` (boolean): Use local Ollama server (convenience for -c model_provider=oss). Requires Ollama running locally. Automatically sets sandbox to workspace-write if not specified.
- `outputLastMessage` (string): Write final Codex message to file path. Useful for CI/CD result capture (Codex CLI v0.95.0+)
- `outputSchema`: JSON Schema path or inline schema to constrain output format (Codex CLI v0.95.0+)
- `personality` (string): Communication style: pragmatic (concise, machine-friendly) or friendly (conversational). Codex CLI v0.94.0+
- `profile` (string): Configuration profile to use from ~/.codex/config.toml
- `prompt` (string, required): Task or question. Use @ to include files (e.g., '@largefile.ts explain').
- `reasoningEffort` (string): Reasoning depth level: low (fast), medium (default), high (complex), xhigh (extra deep)
- `resetSession` (boolean): Clear session context before execution. Starts fresh conversation.
- `responseMode` (string): Response verbosity: "clean" returns only the final answer (default), "full" returns the complete execution log including thinking, tool calls, and agent activity
- `sandbox` (boolean): Quick automation mode: enables workspace-write + on-failure approval. Alias for fullAuto.
- `sandboxMode` (string): Access: read-only, workspace-write, danger-full-access
- `search` (boolean): Enable web search using native --search flag (v0.52.0+). Requires network access - automatically sets sandbox to workspace-write if not specified.
- `sessionId` (string): Session ID for conversation continuity. Enables native Codex resume.
- `skipGitRepoCheck` (boolean): Skip git repository validation. Useful for non-git directories (Codex CLI v0.75.0+)
- `timeout` (number): Maximum execution time in milliseconds (optional)
- `toolOutputTokenLimit` (number): Maximum tokens for tool outputs (100-10,000). Controls response verbosity.
- `workingDir` (string): Working directory for execution
- `yolo` (boolean): ⚠️ Bypass all safety (dangerous)

### `batch-codex` (~244 tokens)

Delegate multiple atomic tasks to Codex for batch processing. Ideal for repetitive operations, mass refactoring, and automated code transformations

Input parameters:

- `disableFeatures` (array): Disable feature flags
- `enableFeatures` (array): Enable feature flags
- `model` (string): Model to use: gpt-5.4, gpt-5.3-codex, gpt-5.2-codex, gpt-5.1-codex-max, gpt-5.1-codex-mini, gpt-5.2
- `oss` (boolean): Use local Ollama server
- `parallel` (boolean): Execute tasks in parallel (experimental)
- `sandbox` (string): Sandbox mode: read-only, workspace-write, danger-full-access
- `search` (boolean): Enable web search for all tasks (activates web_search_request feature)
- `stopOnError` (boolean): Stop execution if any task fails
- `tasks` (array, required): Array of atomic tasks to delegate to Codex
- `timeout` (number): Maximum execution time per task in milliseconds
- `workingDir` (string): Working directory for execution

Output parameters:

- `failed` (number)
- `results` (array)
- `skipped` (number)
- `successful` (number)
- `total` (number)

### `review-changes` (~249 tokens)

Run a code review against the current repository using Codex CLI native review subcommand

Input parameters:

- `base` (string): Review changes against a specific base branch (e.g., "main", "develop")
- `commit` (string): Review the changes introduced by a specific commit SHA
- `model` (string): Model: gpt-5.4, gpt-5.3-codex, gpt-5.2-codex, gpt-5.1-codex-max, gpt-5.1-codex-mini, gpt-5.2. Default: uses Codex's default
- `prompt` (string): Custom review instructions or focus areas (cannot be used with uncommitted=true; use base/commit review instead)
- `timeout` (number): Maximum execution time in milliseconds
- `title` (string): Optional title to display in the review summary
- `uncommitted` (boolean): Review staged, unstaged, and untracked changes (working tree) - cannot be combined with custom prompt
- `workingDir` (string): Working directory to run the review in (passed via -C as a global Codex option)

### `ping` (~19 tokens)

Echo

Input parameters:

- `prompt` (string): Message to echo

### `Help` (~11 tokens)

receive help information

### `version` (~13 tokens)

Display version and system information

Output parameters:

- `codexCli` (string)
- `mcpServer` (string)
- `nodeJs` (string)
- `platform` (string)

### `brainstorm` (~444 tokens)

Generate creative ideas using structured frameworks with domain context and feasibility analysis.

Input parameters:

- `approvalPolicy` (string): Approval: never, on-request, on-failure, untrusted
- `cd` (string): Working directory
- `constraints` (string): Limitations: budget, time, technical, legal, etc.
- `disableFeatures` (array): Disable feature flags
- `domain` (string): Domain: software, business, creative, research, product, marketing, etc.
- `enableFeatures` (array): Enable feature flags
- `existingContext` (string): Background info or previous attempts
- `fullAuto` (boolean): Full automation mode
- `ideaCount` (integer): Number of ideas (default: 12, range: 5-30)
- `includeAnalysis` (boolean): Include feasibility/impact analysis
- `localProvider` (string): Specify which local provider to use (lmstudio or ollama). Automatically enables --oss if not set.
- `methodology` (string): Framework: divergent, convergent, scamper, design-thinking, lateral, auto (default)
- `model` (string): Model: gpt-5.4, gpt-5.3-codex, gpt-5.2-codex, gpt-5.1-codex-max, gpt-5.1-codex-mini, gpt-5.2. Default: gpt-5.4
- `oss` (boolean): Use local Ollama server
- `personality` (string): Communication style: pragmatic (concise, machine-friendly) or friendly (conversational). Codex CLI v0.94.0+
- `prompt` (string, required): Brainstorming challenge or question
- `reasoningEffort` (string): Reasoning depth: low (fast), medium (default), high (complex), xhigh (extra deep)
- `sandboxMode` (string): Access: read-only, workspace-write, danger-full-access
- `search` (boolean): Enable web search for research (activates web_search_request feature)
- `yolo` (boolean): ⚠️ Bypass all safety (dangerous)

### `fetch-chunk` (~66 tokens)

Retrieves cached chunks from a changeMode response. Use this to get subsequent chunks after receiving a partial changeMode response.

Input parameters:

- `cacheKey` (string, required): The cache key provided in the initial changeMode response
- `chunkIndex` (number, required): Which chunk to retrieve (1-based index)

### `timeout-test` (~33 tokens)

Test timeout prevention by running for a specified duration

Input parameters:

- `duration` (number, required): Duration in milliseconds (minimum 10ms)

### `list-sessions` (~56 tokens)

List all active conversation sessions with metadata, or manage sessions

Input parameters:

- `action` (string): Action: list (default), delete (single session), clear (all sessions)
- `sessionId` (string): Session ID for delete action

Output parameters:

- `action` (string)
- `sessions` (array)
- `stats` (object)

### `health` (~43 tokens)

Check Codex CLI and session health status

Input parameters:

- `sessionId` (string): Optional session ID to check specific session health
- `verbose` (boolean): Include detailed diagnostic information

Output parameters:

- `codexCli` (object)
- `features` (object)
- `issues` (array)
- `sessions` (object)
- `status` (string)

### `list-skills` (~42 tokens)

Discover available Codex skills from .agents/skills/ directory

Input parameters:

- `workingDir` (string): Working directory to search for skills. Defaults to resolved working directory.

Output parameters:

- `baseDir` (string)
- `skills` (array)

### `do-act` (~253 tokens)

Execute task via Codex, verify with shell command, auto-fix on failure. Act-Check-Fix loop.

Input parameters:

- `fullAuto` (boolean): Full automation mode
- `localProvider` (string)
- `maxRetries` (integer): Max retry attempts if verify fails. Default: 2
- `model` (string): Model: gpt-5.4, gpt-5.3-codex, gpt-5.2-codex, gpt-5.1-codex-max, gpt-5.1-codex-mini, gpt-5.2. Default: gpt-5.4
- `oss` (boolean): Use local Ollama/LM Studio
- `reasoningEffort` (string): Reasoning depth
- `sandboxMode` (string): Access: read-only, workspace-write, danger-full-access
- `task` (string, required): Task for Codex to execute
- `timeout` (number): Codex timeout per attempt in ms. Default: 10min
- `verify` (object): Shell command to validate result. Triggers act-check-fix loop on failure.
- `workingDir` (string): Working directory

Output parameters:

- `attempts` (number)
- `durationMs` (number)
- `status` (string)
- `steps` (array)

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/x51xxx-codex-mcp-tool/trishchuk-codex-mcp-tool#diagnostics

## Score history

- 2026-08-03: 66
- 2026-08-02: 63
- 2026-08-01: 5
- 2026-07-31: 24
- 2026-07-30: 32
- 2026-07-28: 50
- 2026-07-27: 50

## Links

- npm package: https://www.npmjs.com/package/@trishchuk/codex-mcp-tool
- Socket report: https://socket.dev/npm/package/@trishchuk/codex-mcp-tool
- Repository: https://github.com/x51xxx/codex-mcp-tool
- Changelog RSS feed: https://verifymcp.io/servers/x51xxx-codex-mcp-tool/trishchuk-codex-mcp-tool/changelog.xml
- Changelog JSON feed: https://verifymcp.io/servers/x51xxx-codex-mcp-tool/trishchuk-codex-mcp-tool/changelog.json
- HTML version of this page: https://verifymcp.io/servers/x51xxx-codex-mcp-tool/trishchuk-codex-mcp-tool