AI Workbench MCP

io.github.hrishikesh-thakrePublished by GitHub account hrishikesh-thakrev0.6.0a0Updated Nov 3, 2026

Goose-first MCP server for Workbench-owned acceptance evidence, validation gates, and analytics.

VerifiedSTDIODesktop onlyDeveloper ToolsData & Analytics

Overview

AI-generated overview

Local MCP server that supervises AI coding agents, records run evidence, validates work and produces accept, needs_review or block PR-gate decisions.

What you can do
  • Open and configure runs: start a run, pick a policy pack and select a model.
  • Record agent execution and capture local evidence for the run.
  • Validate a run and write a validation report, then apply a quality gate that writes a revision decision.
  • Analyze runs and render PR-ready reports with accept, needs_review or block outcomes.
When to use it
Use it when you want an auditable acceptance trail for AI coding agent work in a local repository: evidence capture, validation profiles, quality gates and PR-ready reports, especially in Goose, Codex or OpenCode workflows.
Requirements
Runs as a local Python process installed from the PyPI package ai-workbench-mcp; the CLI is invoked as ai-workbench. A project directory must be registered and the local supervisor started; Codex hooks need to be installed and trusted once. No accounts, API keys or environment variables are declared.
Before you install
Public alpha: daemon, Codex hook and OpenCode adapter coverage are alpha mechanisms, and the server checks evidence quality and acceptance readiness rather than proving work is correct; high-risk work still needs human review. It writes run artifacts and reports into the project, and runs/ should not be committed; committed sample evidence must be sanitized.

Installation

In SourceWeft

  1. Open AI Workbench MCP in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

AI Workbench

[License]

AI Workbench supervises AI coding agents, captures evidence, validates work, applies acceptance policy, and produces auditable PR-ready reports.

The PyPI package remains ai-workbench-mcp for this public alpha because the ai-workbench package name is already occupied. The product and CLI are AI Workbench:

bash
pip install ai-workbench-mcpai-workbench --help

Current source metadata targets unpublished ai-workbench-mcp==0.8.0a0. This public alpha consolidates local supervision, evidence capture, validation, acceptance policy, and PR reporting into one product surface.

Public Alpha Warning

The supervisor is the preferred automated evidence path, but daemon, Codex hook, and OpenCode adapter coverage are alpha mechanisms. AI Workbench checks evidence quality and acceptance readiness; it does not prove the work is absolutely correct. High-risk work still requires human review.

Architecture

  • AI Workbench supervisor captures local evidence.
  • AI Workbench validation writes validation_report.json.
  • AI Workbench quality gate writes revision_decision.json.
  • AI Workbench PR/report surfaces render accept, needs_review, or block.

Agent output is a proposal. Workbench accepts evidence.

MCP is the connection protocol. AI Workbench MCP is the tool server. Acceptance is decided by the selected validation profile and quality gate. The agent performs. Workbench accepts. MCP connects them.

Quick Start

Register a project once and start the local supervisor:

bash
pip install ai-workbench-mcpai-workbench supervisor setup --project-dir . --task-type code_changeai-workbench supervisor start

Run Codex, OpenCode, Goose, or another supported local workflow in the project. Then inspect the latest report:

bash
ai-workbench supervisor statusai-workbench reports show latest --project-dir .

Render PR-ready artifacts from a finalized run:

bash
ai-workbench pr-gate --run-dir runs/<run_id>

The canonical local run ledger is:

text
runs/<run_id>/  task_metadata.json  final_prompt.md  model_selection.json  model_output.md  validation_report.json  revision_decision.json  run_log.jsonl  metadata.json  transcript.jsonl  commands.jsonl  workspace/  validation/  artifacts/

validation_report.json and revision_decision.json are the final acceptance authority. Supporting supervisor reports are local evidence, not a substitute for those Workbench artifacts.

Codex Hooks

Install project-local Codex hooks:

bash
ai-workbench setup codex --project-dir . --task-type code_change

Restart Codex or start a new session, open /hooks, review the project hook, and trust it once. Until a hook event is observed, supervisor status reports Codex coverage as configured but unverified.

Goose MCP

AI Workbench still exposes the same MCP tool lifecycle. Register the server with Goose or another MCP host using:

bash
ai-workbench mcp serve

The seven MCP tools remain:

text
workbench_open_runworkbench_select_policy_packworkbench_select_modelworkbench_record_executionworkbench_validate_runworkbench_quality_gateworkbench_analyze_runs

PR Gate

Workbench PR acceptance consumes real Workbench run evidence:

bash
ai-workbench pr-gate \  --run-dir runs/<run_id> \  --out runs/pr_gate/pr_comment.md \  --json-out runs/pr_gate/pr_decision.json

Outcomes are exactly:

  • accept
  • needs_review
  • block

Missing, unreadable, or scaffold-only evidence blocks. A green CI run, uploaded artifact, sticky PR comment, or model self-claim is not acceptance evidence.

Bootstrap Assets

To add starter configs, prompts, recipes, docs, and the GitHub PR-gate workflow to a repository:

bash
ai-workbench bootstrap --target .

The bootstrap keeps runs/ ignored.

Package Demo

For a package-only synthetic demo:

bash
ai-workbench demo --target ./workbench-first-run

This shows accept, needs_review, and block PR-gate outcomes with fixture evidence. It is not a real target-repository acceptance run.

Development

bash
python -m pip install -e ".[dev,publish]"python -m pytest -q -p no:cacheproviderpython -m ruff check . --no-cachepython -m mypy --no-sqlite-cache --no-incrementalai-workbench demo --target runs/package_demo_smokeai-workbench validate --project ai_workbench_mcp --profile scaffold --run-dir runs/scaffold-smoke

Do not commit runs/. Committed sample evidence must be sanitized and live under examples/.

Docs

Recipes:

Sample evidence:

License

Apache-2.0. See LICENSE. MIT-origin attribution for the consolidated Prove It code is retained in NOTICE.

Source: README.md at commit c86ff8f

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.6.0a0LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts