厳格な予算制限と反復回数の制限により、コーディングエージェントによるソフトウェアタスクの実行、検証、およびリカバリを行うためのガバナンスレイヤーを提供します。
別名 Keesan12/martin-loop, martin-loop, @martinloop/mcp
- スター
- 286
- ツール
- —
- ダウンロード数(30日間)
- —
- 最新バージョン
- 0.9.0
- 最終コミット
- 2026年11月4日
- ライセンス
- Apache-2.0
Martin Loop MCPとは
厳格な予算制限と反復回数の制限により、コーディングエージェントによるソフトウェアタスクの実行、検証、およびリカバリを行うためのガバナンスレイヤーを提供します。
プロジェクトのREADMEから生成された概要
Martin Loop MCP server について
Martin Loopは、コーディングエージェントを使用する開発者が、厳格な境界内でソフトウェアタスクを管理するためのガバナンスレイヤーを提供します。これはエージェントの周囲に配置される制御システムとして機能し、定義された成功基準と検証ステップを通じて、ジョブが集中力を維持し、責任ある動作を行うことを保証します。プリフライトチェック、制御された実行、実行後の分析を含む構造化されたライフサイクルを実装することで、エージェントが目的から逸脱したり、過剰なリソースを消費したりすることを防ぎます。
ユーザーは、--budget-usd フラグによる支出上限の設定や、--max-iterations による試行回数の制限を通じて、コストと実行制限を管理できます。このシステムはモデルに依存せず、さまざまなランタイムで動作するため、ジョブの定義を維持したままワーカーを切り替えることが可能です。主にnpxコマンドまたはnpmによるグローバルインストールを通じてローカルで使用されますが、MARTIN_API_TOKEN を使用してエビデンスをダッシュボードに送信する、オプションのホスト同期機能も提供しています。開発者は、auditコマンドを実行することで、既存のエージェントの支出を監査し、修正とリトライのループにどれだけの費用がかかったかを確認できます。
できること
- 1厳格なタスク予算の強制開発者は、特定の金額を設定することで、AIエージェントがコーディングタスクに費やす金額を制限できます。
- 2エージェントのリトライ試行回数の制限ジョブに許可される総反復回数を上限に設定することで、エージェントが無限ループに陥るのを防ぐことができます。
- 3コード完了の自動検証AIクライアントは、特定のコマンドを使用して、タスクが終了する前に必要な成功基準を満たしていることを証明できます。
- 4エージェントの支出履歴の監査開発者は、過去のエージェントセッションにどれだけの費用がかかったかを分析し、非効率なリトライループを特定できます。
プロジェクトの README から作成されています。すべてのユースケースは、その引用と照らし合わせて確認されています。
Martin Loop MCPのインストール方法
フォーマット検証日 2026年11月10日Cursor に追加Gooseに追加Claude CodeにMartin Loop MCPをインストール
$ claude mcp add martin-loop -- npx -y @martinloop/mcpCursorにMartin Loop MCPをインストール
~/.cursor/mcp.json
{
"mcpServers": {
"martin-loop": {
"command": "npx",
"args": [
"-y",
"@martinloop/mcp"
]
}
}
}VS CodeにMartin Loop MCPをインストール
.vscode/mcp.json
{
"servers": {
"martin-loop": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@martinloop/mcp"
]
}
}
}VS Code CLIにMartin Loop MCPをインストール
$ code --add-mcp '{"name":"martin-loop","type":"stdio","command":"npx","args":["-y","@martinloop/mcp"]}'Claude DesktopにMartin Loop MCPをインストール
claude_desktop_config.json
{
"mcpServers": {
"martin-loop": {
"command": "npx",
"args": [
"-y",
"@martinloop/mcp"
]
}
}
}CodexにMartin Loop MCPをインストール
$ codex mcp add martin-loop -- npx -y @martinloop/mcpCodex config.tomlにMartin Loop MCPをインストール
~/.codex/config.toml
[mcp_servers.martin-loop] command = "npx" args = ["-y", "@martinloop/mcp"]
Gemini CLIにMartin Loop MCPをインストール
$ gemini mcp add -s user martin-loop npx -- -y @martinloop/mcpGemini settings.jsonにMartin Loop MCPをインストール
~/.gemini/settings.json
{
"mcpServers": {
"martin-loop": {
"command": "npx",
"args": [
"-y",
"@martinloop/mcp"
]
}
}
}自動チェック
実行日 2026年11月12日完全なレポート →自動チェックに合格しました。これはセキュリティ監査ではありません。コードを読み、サーバーに必要最小限のアクセス権を与えてください。 チェックの仕組み
- ソースリポジトリがリンクされており、アクセス可能合格
- オープンソースライセンス (OSI)合格
- 過去1年以内にコミットまたはリリースあり合格
- パッケージ名が有名なパッケージに酷似していない合格
- リモートエンドポイントがHTTPSを使用未チェック
- 破壊的なツールがラベル付けされている未チェック
- モデルへの隠れた指示がない合格
- パッケージおよびその依存関係に重大な勧告なし合格
README
GitHubで見るMartinLoop
Give coding agents more work. Watch them less. Ship more.
MartinLoop lets you give coding agents real software jobs without babysitting every step.
Your coding agent still writes the code. MartinLoop keeps the job focused, bounded, and accountable — with limits, stop conditions, verification, recovery, and a clear outcome when the run ends.
Trust your coding agents with real work.
Built from thousands of agent runs where the problem was not just intelligence — it was whether you could give the agent a real job and confidently walk away.
Measure your loop tax: npx -y martin-loop@latest audit
Get started: npx -y martin-loop@latest start
Try the demo: npx -y martin-loop@latest demo
MartinLoop is part of the NVIDIA Inception program.
Start Here
Audit first — run npx -y martin-loop@latest audit to see how much API-equivalent Claude Code spend went into fix-and-retry loops. Add --offline for zero network access or --share for a shareable summary.
Install / onboard — run npx -y martin-loop@latest start, or install it globally with npm install -g martin-loop@latest.
Governed run — define an objective, verifier, budget, and iteration cap with martin run.
Verifier — completion requires fresh verifier evidence bound to the active run and workspace. A configured verifier proves only the checks it runs; VERIFIED is not a claim that the code is bug-free or automatically safe to merge.
Budget — set a hard spend ceiling with --budget-usd and an attempt ceiling with --max-iterations.
Receipts — inspect the latest result with martin dossier --latest and validate stored integrity with martin runs verify --latest.
Hosted sync (optional) — governed work is local-first. Configure MARTIN_API_TOKEN and MARTIN_TELEMETRY_ENDPOINT, then use martin sync status and martin sync flush to send preserved evidence to a dashboard later. See the quickstart.
MCP — install @martinloop/mcp@latest in a supported host or generate host configuration with martin mcp print-config.
Documentation — continue with the quickstart, CLI reference, or MCP setup.
When --model is provided, MartinLoop passes it through unchanged. Without --model, the authenticated host runtime chooses its own default. MartinLoop does not inject a hidden fallback model.
For AI Agents
MartinLoop is the system around the coding job.
The coding agent still writes the code. MartinLoop keeps the job, limits, verification, recovery, outcome, and history consistent around the work.
Use MartinLoop when a coding task needs:
- clear job and success criteria
- preflight readiness before agent spend
- hard budget and iteration limits
- allowed or denied edit scope
- controlled retries and stop conditions
- independent verifier commands
- recovery and rollback evidence
- a finite completion decision
- run receipts, dossiers, history, failure classification, and post-run analysis
- one consistent workflow across supported coding-agent runtimes
Canonical lifecycle:
DEFINE
-> PREFLIGHT
-> CONTROL
-> VERIFY
-> RECOVER
-> PROVE
-> ANALYZE
The product-level flow is Definition of Done -> Controlled Run -> Verified Handoff.
For machine-readable context start with llms.txt, llms-full.txt, and MartinLoop for AI Agents.
One System Around the Run
Teams should not need to stitch together a separate script or point tool for every part of coding-agent execution. MartinLoop connects the control path around the agent from preflight through post-run evidence.
| Stage | MartinLoop role |
|---|---|
| Define | Capture the objective, verifier, budget, scope, and finish line. |
| Preflight | Check readiness and required workflow evidence before agent spend. |
| Control | Enforce budgets, attempts, path boundaries, policy, and stop conditions while the coding agent works. |
| Verify | Run configured checks and bind the evidence to the active run and workspace. |
| Recover | Preserve recovery and rollback state when another attempt or human review is required. |
| Prove | Produce the authoritative VERIFIED, STOPPED, or NEEDS REVIEW handoff plus receipts. |
| Analyze | Inspect run history, cost provenance, failure classes, dossiers, and shareable evidence after execution. |
MartinLoop does not replace Git, GitHub, CI, dedicated security scanners, observability platforms, code review, or the coding agent itself. It gives those workflows one governed execution record to inspect.
Change the worker. Keep the job.
MartinLoop is model-agnostic by design. The job, limits, verification, recovery, and outcome are separate from the worker underneath.
Native coding-agent CLIs and OpenAI-compatible model runtimes use different execution paths, so MartinLoop only describes a worker as fully supported when that execution path has been publicly released and validated.
Why MartinLoop
AI coding agents can create more software than ever. The problem is trusting them with bigger jobs without turning yourself into their full-time manager.
Agents drift. They retry. They break unrelated things. They say "done" too early. And a run that was supposed to save time can create even more work to review.
MartinLoop keeps the job accountable outside the agent itself: clear boundaries, hard limits, configured checks, recovery information, and an explicit outcome.
The goal is simple:
Give the agent more responsibility without giving it unlimited freedom.
Why Teams Adopt MartinLoop
- Less babysitting — give agents larger jobs without watching every step.
- Clear boundaries — keep scope, attempts, and spend inside the limits you set.
- Checks before "done" — configured verification decides whether the evidence passed.
- Understand failures — know why the run stopped instead of digging through logs.
- Recover cleanly — preserve enough evidence to retry, repair, roll back, or hand off.
- One workflow across agents — keep the job consistent as the worker underneath changes.
More software. Less supervision.
See Your Loop Tax
Before changing your workflow, measure it:
npx -y martin-loop@0.9.0 audit
MartinLoop reads Claude Code session history locally and reports how much API-equivalent agent spend happened inside fix-and-retry loops, how often verifier commands failed, the longest retry chain, and sessions that ended red. Session contents stay local. Use --offline to disable even the public pricing lookup, and --share to create a screenshot-ready SVG plus Markdown summary.
Then put limits around the next run:
npx -y martin-loop@0.9.0 enable --engine codex
npx -y martin-loop@0.9.0 run "<task>" --verify "npm test" --budget-usd 2 --max-iterations 3
2-Minute Install Path
If MartinLoop detects more than one supported coding-agent runtime, save the one you want to use before the first governed run. For example:
npx -y martin-loop@latest enable --engine codex
Replace codex with claude or gemini when that is your intended runtime.
npx -y martin-loop@latest start
npx -y martin-loop@latest demo
cd martin-loop-demo
npm install
npx -y martin-loop@latest run "Summarize the demo workspace and prove tests still pass" --verify "npm test" --budget-usd 2 --max-iterations 1
Quick Start
Try MartinLoop in a disposable demo workspace:
npx -y martin-loop@latest enable --engine codex
npx -y martin-loop@latest start
npx -y martin-loop@latest demo
npx -y martin-loop@latest --version
cd martin-loop-demo
npm install
npx -y martin-loop@latest run "Summarize the demo workspace and prove tests still pass" --verify "npm test" --budget-usd 2 --max-iterations 1
npx -y martin-loop@latest dossier --latest
npx -y martin-loop@latest share --latest
Optional global install:
npm install -g martin-loop
martin-loop --version
If this flow is useful, open an issue with feedback so we can keep improving the public experience.
start prints the first-run guided path. run auto-checks doctor, session-start, and preflight, then executes when the environment is ready. Use --proof only when you intentionally want an explicit no-spend lane.
Inspect-first flow:
npx -y martin-loop@latest enable --engine codex
npx -y martin-loop@latest doctor
npx -y martin-loop@latest session-start
npx -y martin-loop@latest preflight "Summarize the demo workspace and prove tests still pass" --verify "npm test"
share --latest writes three files into the selected run directory under share/: run-receipt.json, run-receipt.md, and proof-card.svg.
Release notes for MartinLoop 0.6.5: MartinLoop 0.6.5.
Release notes for MartinLoop 0.6.6: MartinLoop 0.6.6.
Release notes for MartinLoop 0.6.7: MartinLoop 0.6.7.
Release notes for MartinLoop 0.6.8: MartinLoop 0.6.8.
Release notes for MartinLoop 0.7.0: MartinLoop 0.7.0.
Release notes for MartinLoop 0.7.1: MartinLoop 0.7.1.
Release notes for MartinLoop 0.8.0: MartinLoop 0.8.0.
Release notes for MartinLoop 0.8.1: MartinLoop 0.8.1.
Release notes for MartinLoop 0.8.2: MartinLoop 0.8.2.
Release notes for MartinLoop 0.8.3: MartinLoop 0.8.3.
Release notes for MartinLoop 0.8.4: MartinLoop 0.8.4.
Release notes for MartinLoop 0.8.5: MartinLoop 0.8.5.
Release notes for MartinLoop 0.8.6: MartinLoop 0.8.6.
Release notes for MartinLoop 0.8.7: MartinLoop 0.8.7.
Release notes for MartinLoop 0.8.8: MartinLoop 0.8.8.
Release notes for MartinLoop 0.8.9: MartinLoop 0.8.9.
Release notes for MartinLoop 0.9.0: MartinLoop 0.9.0.
Swarm Mode Demo
Run the deterministic 15-agent launch-board demo without provider calls or spend:
npx -y martin-loop@0.9.0 demo --swarm
That default result is local simulation evidence only: it is non-governed, is not persisted to the Swarm run store, and does not evaluate signed receipt integrity. Inspect simulation output through the demo --swarm human view or demo --swarm --json, plus the local files created under --dir. Do not use swarm status, swarm inspect, swarm dossier, swarm verify, or swarm share for the default simulation output.
To exercise the persisted provider-free governance path, use:
npx -y martin-loop@0.9.0 demo --swarm --governed --agents 15 --max-concurrency 5 --budget-usd 0
An approved live Codex qualification uses the same persisted route with --live, an explicit positive budget, and a clean repository baseline. Persisted modes use --cwd and --runs-dir; they reject the simulation-only --dir and --force flags. Live failures exit nonzero and never fall back to simulation.
Persisted demo runs and runs created by swarm plan plus swarm run are valid inputs to swarm status, swarm inspect, swarm dossier, swarm verify, and swarm share. A stored run is not automatically verified: rely on the verifier and receipt-integrity result, not the command route alone.
Model and Engine Support
MartinLoop governs the job independently of the coding worker.
- Native agent CLIs: Codex, Claude Code, Gemini CLI
- OpenAI-compatible endpoints: Kimi K2, NVIDIA Nemotron, DeepSeek, Qwen/Qwen Coder, Mistral/Codestral, OpenRouter/Together/Fireworks routes, Ollama, LM Studio, llama.cpp, and other compatible endpoints
- Use
--engine openaifor OpenAI-compatible model endpoints.
The worker changes; MartinLoop's budget, scope, verifier, receipt, and integrity contract does not.
More detail: Model and engine support
The Run From Start to Handoff
MartinLoop's terminal presentation is built around the governed lifecycle, not around a single verifier command.
Governed Run Plan shows the configured finish line before work starts, including the task, budget posture, verifier plan, scope, and execution boundaries.
Controlled Run keeps the coding agent working inside those boundaries while MartinLoop tracks attempts, cost, stop conditions, and recovery state.
Verified Handoff closes the loop with one authoritative outcome:
VERIFIEDwhen the configured evidence supports the Definition of DoneSTOPPEDwhen a configured hard boundary ends the runNEEDS REVIEWwhen completion cannot be established from the available evidence
The handoff can include verifier steps, scope state, attempt count, cost provenance, unresolved evidence, recovery state, receipt integrity, and the next safe action. The exact fields depend on what the run actually established.
Visual Proof
MartinLoop turns an AI coding run into an inspectable execution record: budget used, verifier result, changed files, rollback evidence, and final receipt.
Ungoverned agents can retry until cost and scope drift. MartinLoop adds budget caps, verifier gates, and audit evidence so the run has a clear stop condition.
MartinLoop Arcade
Long governed runs do not have to mean staring at a spinner. In an interactive terminal, MartinLoop Arcade can be offered while the coding agent continues working in the background.
Arcade is presentation-only. It cannot change the agent, budget, verifier, policy decision, run outcome, or receipt evidence. It stays out of JSON, CI, non-interactive, and other machine-readable execution paths.
Use --arcade to offer Arcade immediately for a supported interactive run, or --no-arcade to suppress it for that run.
Proof Receipts
Proof receipts are local share bundles for governed AI coding runs. They show the task, spend, budget, verifier result, receipt integrity, and any evidence boundary that should not be rounded into confidence.
This real governed run spent $0.51 against a $3.00 budget. The verifier passed and the receipt integrity was signed, but the proof stayed at EVIDENCE_BOUNDARY because rollback evidence was not recorded.
Generate your own receipt after a governed run:
npx -y martin-loop@latest run "Summarize the demo workspace and prove tests still pass" --proof --verify "npm test"
npx -y martin-loop@latest runs verify --latest
npx -y martin-loop@latest share --latest
Example receipt files: Markdown and JSON.
Verify the Public Release Yourself
Use this lane from a clean temp directory to verify the public CLI flow exactly as shipped:
npx -y martin-loop@0.9.0 enable --engine codex
npx -y martin-loop@0.9.0 --version
npx -y martin-loop@0.9.0 start
npx -y martin-loop@0.9.0 demo
cd martin-loop-demo
npm install
npx -y martin-loop@0.9.0 run "Summarize the demo workspace and prove tests still pass" --verify "npm test" --budget-usd 2 --max-iterations 1 --json
npx -y martin-loop@0.9.0 dossier --latest --json
npx -y martin-loop@0.9.0 share --latest --json
For deterministic installs, pin the package line (martin-loop@0.9.0) or use martin-loop@latest. Plain npx martin-loop can resolve a stale local cache on some machines.
Expected share bundle outputs:
share/run-receipt.jsonshare/run-receipt.mdshare/proof-card.svg
See It In Action
The point is not that every governed run is always cheaper. The point is that every run becomes inspectable and enforceable: budget policy, verifier result, stop reason, and evidence are explicit.
For a deterministic public repro lane, use the benchmark workspace and compare governed execution to unbounded retry behavior:
npx martin-loop bench --suite under-3-challengenpx martin-loop bench --suite ralphy-engineering-50
Ralph-Style Loops
A Ralph-style loop is the failure mode where an AI coding agent keeps trying without knowing when continuing is unsafe, uneconomical, or unlikely to succeed.
MartinLoop keeps the useful part of the loop, then adds brakes:
- stop before budget overspend
- classify unsafe or invalid actions before execution
- write an audit record for every attempt
- preserve rollback and verifier evidence for review
- reduce runaway context growth with compact run summaries
Failure Taxonomy (13 Runtime Classes)
When a coding agent fails, "something went wrong" is not useful enough.
MartinLoop classifies runtime failures into 13 canonical classes so a failed run can tell you what happened, what needs attention, and what future runs can learn from.
See the canonical table: Failure Taxonomy (13 Runtime Classes).
What It Does
- Budget caps stop the next attempt before a configured USD, token, or iteration limit is exceeded.
- Verifier gates require a real check, such as
npm test, before a run can count as complete. - Policy checks block unsafe verifier commands, risky path changes, and secret-like task inputs before execution.
- Failure classification uses canonical runtime classes for triage and reporting. See Failure Taxonomy (13 Runtime Classes).
- Run receipts capture stop reason, verifier evidence, budget posture, integrity state, and the next safe action.
martin share --latestturns the latest governed run into a local share bundle with a redacted JSON receipt, Markdown recap, and proof-card SVG.- MCP integration gives hosts one write-capable execution entrypoint plus richer planning, inspection, and review helpers.
How It Works
| Layer | Purpose |
|---|---|
| Task contract | Objective, verifier plan, repo root, allowed paths, denied paths, acceptance criteria, workspace, project, and budget. |
| Policy and budget | Defaults come from martin.config.yaml; CLI flags can override them. Budget preflight blocks attempts that would exceed policy. |
| Agent adapters | Claude CLI, Codex CLI, Gemini CLI, and direct-provider adapters normalize execution results. |
| Safety and verification | Scope checks, verifier command checks, prompt integrity, and grounding decide whether work can continue. |
| Persistence | JSONL run records, evidence summaries, and repo-backed artifacts make every run inspectable later. Each loop record is locally signed (HMAC, per-runs-root key) and dossier/runs get/runs verify/challenge/badge report an integrity verdict (verified / tamper_detected / unsigned) so post-hoc edits to a record are detectable, not just inspectable. |
Trust Boundaries
- Cost and token outputs always include provenance (
actual,calculated,estimated, orunavailable). - For Codex specifically, MartinLoop reports authoritative usage only when the host exposes it; otherwise MartinLoop labels usage as estimated and avoids presenting it as settled accounting.
- Receipt integrity must be
verifiedbefore a run is treated as trustworthy evidence for external review.
Martin Loop MCPのよくある質問
Martin Loop MCPをClaude Codeに追加するには
ターミナルでclaude mcp add martin-loop -- npx -y @martinloop/mcpを実行します。すべてのプロジェクトで利用するには--scope userを追加します。
Martin Loop MCPをCursorに追加するには
~/.cursor/mcp.jsonを開き、上記のインストールセクションに示すCursor設定を使ってmcpServersの下にサーバーを追加します。
コーディングエージェントの実行コストを制限するにはどうすればよいですか?
実行時に --budget-usd フラグを使用して、支出の上限を設定できます。
MartinLoopはコーディングエージェント自体に代わるものですか?
いいえ、コードを書くのはコーディングエージェントです。MartinLoopは、その作業に対する制御パス、制限、および検証を提供します。
エージェントのリトライループにどれくらいの費用がかかっているか監査できますか?
はい、auditコマンドを実行することで、修正とリトライのループにどれだけのAPI相当の費用がかかったかを確認できます。
検証プロセスは、コードにバグがないことを保証するものですか?
いいえ、設定された検証器は、プログラムされた特定のチェックがパスしたことを証明するのみです。