What makes TRAECNclaw MCP notable?

1.2K downloads. Trust evidence available

How should TRAECNclaw MCP be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

TRAECNclaw MCP

Operate TraeCN through TRAECNclaw's focused stdio MCP tools. Use when an MCP-capable agent must send or stop work, choose a workspace/model/mode/conversation, inspect or change Trae settings through its UI, manage conversations, or resolve an exceptional question or command approval. Queueing, waiting, recovery, notifications, routine non-command questions, and keep/revert gates are gateway-managed.

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.2k

Updated

Oct 11, 2026

Version

0.6.0

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.2K downloads reported by the source. Last updated 10/11/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 11, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 11, 2026
Adoption signal: 1.2K downloadsadoption · observed Oct 11, 2026
Latest release: 0.6.0release · observed Aug 24, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s179fnaja3kqc64tnfyq6b5zb583p7d8:traecnclaw-mcp

Install using `clawhub skill install s179fnaja3kqc64tnfyq6b5zb583p7d8:traecnclaw-mcp` in an isolated environment before connecting it to live workloads.
No published capability contract is available yet, so validate auth and request/response behavior manually.
Review the upstream CLAWHUB listing at https://clawhub.ai/luckycat133/traecnclaw-mcp before using production credentials.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-luckycat133-traecnclaw-mcp/snapshot"

Documentation

CLAWHUB

148,296 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: traecnclaw-mcp
description: Operate TraeCN through TRAECNclaw's focused stdio MCP tools. Use when an MCP-capable agent must send or stop work, choose a workspace/model/mode/conversation, inspect or change Trae settings through its UI, manage conversations, or resolve an exceptional question or command approval. Queueing, waiting, recovery, notifications, routine non-command questions, and keep/revert gates are gateway-managed.
license: MIT-0
metadata:
author: TRAECNclaw
version: "0.6.0"
openclaw:
requires:
bins:
- node
primaryEnv: TRAECN_GATEWAY_TOKEN
envVars:
- name: TRAECN_GATEWAY_HOST
required: false
description: Optional gateway host override; keep loopback unless remote access is secured.
- name: TRAECN_GATEWAY_PORT
required: false
description: Optional gateway port override.
- name: TRAECN_GATEWAY_TOKEN
required: false
description: Optional gateway secret; never print, log, or persist it.
- name: TRAECN_MCP_CLIENT_ID
required: false
description: Stable client identifier for legacy MCP task notifications.
---

# TRAECNclaw MCP

Compose focused tools around user intent. Do not call preflight, wait, poll,
event acknowledgement, recovery, cleanup, proof, process, or generic command
operations from the agent loop; the gateway owns those mechanics.

## Setup

1. Install the matching TRAECNclaw server from its source checkout or release server archive. The launcher checks the repository, an installed `traecnclaw` package, then `traecnclaw-mcp` on `PATH`; it never executes a JavaScript path selected by an environment variable.
2. Run `npm run doctor` in a source checkout. Then run `npm run --silent setup:mcp` from the checkout, or `node scripts/setup-mcp.js` from this Skill directory, to validate the server and print a resolved client entry. Use `--output FILE` to create a new mode-0600 file atomically; the command never overwrites an existing file. The JSON templates under [assets](assets) remain manual fallbacks.
3. Start the gateway with `npm run start:gateway`, then reconnect the MCP client.
4. Use mock mode only for development; never cite mock output as live TraeCN evidence.

## Choose only necessary context

- Call `traecn_open_workspace` only when the requested folder is not already open.
- Call `traecn_list_models` when the available choices matter; then call `traecn_select_model` once.
- Call `traecn_select_mode` only when the task requires a different Solo/IDE mode.
- Call `traecn_list_conversations` to read the current mode and stable conversation IDs.
- Use `traecn_create_conversation`, `traecn_select_conversation`, or `traecn_delete_conversation` for exactly one conversation operation.

These tools change TraeCN context. Never repeat workspace paths in the model
message.

## Send and track work

Call `traecn_send_message` with the exact message and, when concurrency makes
the target ambiguous, an explicit `

_meta.json

{
  "ownerId": "kn73ym4c6fp5b2vaa22t84v5fn83pt1p",
  "slug": "traecnclaw-mcp",
  "version": "0.6.0",
  "publishedAt": 1787566992010
}

references/mcp-call-examples.json

{
  "calls": [
    {
      "name": "open_workspace",
      "tool": "traecn_open_workspace",
      "arguments": { "path": "/absolute/path/to/repository" }
    },
    {
      "name": "discover_models",
      "tool": "traecn_list_models",
      "arguments": {}
    },
    {
      "name": "select_model",
      "tool": "traecn_select_model",
      "arguments": { "model": "GLM-5.2" }
    },
    {
      "name": "send_repository_review",
      "tool": "traecn_send_message",
      "arguments": {
        "message": "Review the current repository. Report findings first, ordered by severity."
      }
    },
    {
      "name": "explicit_task_check",
      "tool": "traecn_get_task",
      "arguments": { "taskId": "TASK_ID" }
    },
    {
      "name": "diagnostic_task_trace",
      "tool": "traecn_get_task",
      "arguments": { "taskId": "TASK_ID", "detailLevel": "trace" }
    },
    {
      "name": "list_general_settings",
      "tool": "traecn_list_settings",
      "arguments": { "section": "通用" }
    },
    {
      "name": "toggle_auto_save",
      "tool": "traecn_set_setting_toggle",
      "arguments": { "section": "通用", "label": "Auto Save", "enabled": true }
    },
    {
      "name": "select_conversation",
      "tool": "traecn_select_conversation",
      "arguments": { "conversationId": "solo:UmV2aWV3IEFQSQ" }
    },
    {
      "name": "stop_visible_untracked_generation",
      "tool": "traecn_stop_generation",
      "arguments": {
        "conversationId": "solo:UmV2aWV3IEFQSQ",
        "acknowledgeUntrackedWork": true,
        "reason": "The user asked to stop the generation visible in this conversation"
      }
    },
    {
      "name": "delete_inactive_conversation",
      "tool": "traecn_delete_conversation",
      "precondition": "Use only when the current user request explicitly names or unambiguously identifies this conversation for deletion; do not infer deletion from cleanup or maintenance intent.",
      "warning": "Deletion is permanent, and the gateway cannot restore the conversation. Re-read the listed conversation ID and exact title before sending this call.",
      "arguments": {
        "conversationId": "solo:T2xkIHJldmlldw",
        "expectedTitle": "Old review",
        "acknowledgePermanentDeletion": true
      }
    },
    {
      "name": "answer_exceptional_question",
      "tool": "traecn_answer_question",
      "arguments": { "answer": "Review the current workspace changes" }
    },
    {
      "name": "approve_exact_exceptional_command",
      "tool": "traecn_decide_approval",
      "arguments": {
        "decision": "approve",
        "expectedCommand": "git push origin main",
        "acknowledgeRisk": true,
        "reason": "The user explicitly requested publishing the current branch"
      }
    }
  ]
}

references/mcp-surface.md

# MCP surface

TRAECNclaw exposes one focused Agent surface. There are no tool profiles.

## Protocol and transport

- MCP `2026-07-28`: stateless per-request `_meta`, `server/discover`, typed
  results, structured tool content, server identity metadata, and cache hints.
- MCP `2025-11-25` and `2024-11-05`: legacy `initialize` compatibility.
- stdio: one compact JSON-RPC object per line. `Content-Length` framing is
  accepted only as a migration compatibility path.

MCP contract version 5 below identifies this Skill's frozen 20-tool surface;
it is not an MCP protocol revision. Legacy clients receive compact
`notifications/message` task events. For MCP `2026-07-28`, the server advertises
the separately negotiated `io.modelcontextprotocol/tasks` extension. An
opted-in `traecn_send_message` returns the durable gateway ID as a Task;
`tasks/get`, `tasks/update`, and `tasks/cancel` operate that same identity, and
`subscriptions/listen` can stream exact-ID `notifications/tasks` snapshots.
For a legacy client ID's first connection, already-retained events are
baselined without delivery; a returning ID resumes only events newer than its
durable acknowledgement. Unrelated request metadata such as `progressToken`
does not change the negotiated legacy protocol era.
Modern retries derive a stable idempotency key from the request identity and
arguments. The gateway retains the corresponding terminal fingerprint and
response across restart for a configurable 24-hour default TTL, so the same
request reuses the original task while different work cannot collide.
Clients that do not opt in retain the ordinary complete result and use
`traecn_get_task` only for an intentional read. The extension is upstream-draft
and does not change the 20-tool contract.

| Tool | Required input | Gateway route |
| --- | --- | --- |
| `traecn_send_message` | `message` | `POST /api/tasks/submit` |
| `traecn_get_task` | `taskId`; optional `detailLevel` (`result` or `trace`) | `GET /api/task/{taskId}` |
| `traecn_cancel_task` | `taskId` | `POST /api/task/{taskId}/cancel` |
| `traecn_stop_generation` | `conversationId`, `acknowledgeUntrackedWork`, `reason` | `POST /api/trae/stop-generation` |
| `traecn_open_workspace` | `path` | `POST /api/open-project` |
| `traecn_list_models` | none | `GET /api/models` |
| `traecn_select_model` | `model` | `POST /api/switch-model` |
| `traecn_select_mode` | `mode` | `POST /api/switch-mode` |
| `traecn_list_setting_sections` | none | `GET /api/settings/sections` |
| `traecn_list_settings` | `section` | `GET /api/settings/read` |
| `traecn_list_setting_options` | `section`, `label` | `GET /api/settings/options` |
| `traecn_set_setting_toggle` | `section`, `label`, `enabled` | `POST /api/settings/set` |
| `traecn_select_setting_option` | `section`, `label`, `value` | `POST /api/settings/set` |
| `traecn_set_setting_text` | `section`, `label`, `text` | `POST /api/settings/set` |
| `traecn_list_conversations` | none | `GET /api/conversations` |
| `traecn_cr

references/mcp-tool-contracts.json

{
  "source": "mcp-server.js",
  "contractVersion": 5,
  "toolCount": 20,
  "tools": [
    {
      "name": "traecn_send_message",
      "description": "Submit one user instruction to TraeCN as durable background work and return its stable taskId immediately; the gateway owns queueing, retries, recovery, and result persistence. Use this for every delegated coding request, optionally targeting an existing conversation via conversationId. Do not send follow-up messages to probe progress — track the returned task instead.",
      "inputSchema": {
        "type": "object",
        "properties": {
          "message": {
            "type": "string",
            "description": "The exact instruction to execute."
          },
          "conversationId": {
            "type": "string",
            "description": "Stable identifier returned by traecn_list_conversations."
          }
        },
        "required": [
          "message"
        ],
        "additionalProperties": false
      },
      "metadata": {
        "layer": "task",
        "risk": "delegate"
      }
    },
    {
      "name": "traecn_get_task",
      "description": "Read one known gateway task by ID: status plus final result, or the full execution trace with detailLevel 'trace' for diagnosis. Call once when notified of completion or once later when resuming after a disconnect. Not a progress primitive — do not poll in a tight loop.",
      "inputSchema": {
        "type": "object",
        "properties": {
          "taskId": {
            "type": "string",
            "description": "Stable task identifier returned by traecn_send_message."
          },
          "detailLevel": {
            "type": "string",
            "enum": [
              "result",
              "trace"
            ],
            "description": "result returns only the final answer (default); trace includes the full visible Trae execution process."
          }
        },
        "required": [
          "taskId"
        ],
        "additionalProperties": false
      },
      "metadata": {
        "layer": "task",
        "risk": "read"
      }
    },
    {
      "name": "traecn_cancel_task",
      "description": "Cancel one gateway-tracked task by ID. The gateway stops shepherding the work at its next control point and records the cancellation; results already produced remain in task history. Returns the task identity and post-cancellation status.",
      "inputSchema": {
        "type": "object",
        "properties": {
          "taskId": {
            "type": "string",
            "description": "Stable task identifier returned by traecn_send_message."
          }
        },
        "required": [
          "taskId"
        ],
        "additionalProperties": false
      },
      "metadata": {
        "layer": "task",
        "risk": "delegate",
        "destructive": true
      }
    },
    {
      "name": "traecn_stop_generation",
      "description": "Stop only the generation currently visible in the named acti

Github ReposUpdated 1d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.