Mmcp.market

Martin Loop MCP server

by Keesan12·io.github.Keesan12/martin-loop·v0.6.1
No reviews yet

One system around coding agents to run, verify, recover, and hand off software work.

A92/100grade A
Write a reviewInstall
What users say
No reviews yet
Be the first
Safety scan
A92/100

full report

Adoption
Established

190 stars366 downloads/wk

Reviews

Write one

Nobody has reviewed Martin Loop yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Martin Loop tools (24, 6 write)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • martin_arcade

    Open the presentation-only MartinLoop Arcade when the host supports MCP Apps. The Arcade cannot change governed execution or evidence.

  • martin_arcade_status

    Read the read-only evidence projection used by the Arcade view. Provide loopId for one exact run or omit it to load the latest run. Use only for Arcade rendering; use martin_status for budget pressure or martin_run_dossier for full evidence. This tool reads persisted evidence and never changes run state.

  • martin_cancelwrite action

    Write a durable cancellation receipt for one canonical MartinLoop run. Provide exactly one selector: file, loopId, or latest; runsDir changes the run-store root, while reason and requestedBy add audit context. Use when work must be abandoned, not temporarily held. This records a request and does not kill a process; use martin_pause for a reversible hold.

  • martin_continue

    Record a durable continue or resume request for a Martin run after a human pause or approval checkpoint.

  • martin_create_prwrite action

    Create or preview a GitHub PR with a MartinLoop dossier body. Use execute=true to actually call gh.

  • martin_doctor

    Read-only environment, engine, workspace, and run-store diagnostics for MartinLoop. Use first for software work, fresh installs, suspicious state, or before retries. Do not use as proof that a task is complete. Next: call martin_estimate or martin_triage_runs depending on whether this is new work or failed prior work.

  • martin_dossier

    Read a formatted evidence summary for one MartinLoop run. Provide exactly one selector: file, loopId, or latest; runsDir changes the run-store root. Set format to json, md, or github-pr; json is the default. Use after martin_run, before merge or release claims, or when sharing what happened. Do not use as a substitute for missing verifier evidence; use martin_run_dossier for the full structured re

  • martin_estimate

    Read-only cost, route, and pre-work burn estimate for a software objective. Use before planning/preflight when a change may spend agent time or exceed budget. Do not use for casual questions or as permission to execute. Next: call martin_plan if the estimate is acceptable, or ask for consent if budget/risk is high.

  • martin_eval

    Grade a MartinLoop run across six dimensions — task completion, verifier health, diff discipline, regression risk, security risk, and reviewability — and return a scored merge-readiness verdict.

  • martin_get_attempt

    Load one Martin attempt summary with artifact directory references for a canonical run.

  • martin_get_runwrite action

    Load one Martin run and return its budget, cost, verification, artifact, and canonical path summary.

  • martin_get_verification_results

    Load structured verification evidence for a MartinLoop run — verifier commands, pass/fail outcomes, and contradiction signals — from stored loop events and ledger entries.

  • martin_inspect

    Summarise Martin Loop run records from a saved loop file or run-store directory.

  • martin_list_runs

    List recent Martin runs from the run store with lightweight filters for status, lifecycle, engine metadata, and recency.

  • martin_logs

    Read recent MartinLoop events, ledger entries, and operator control receipts for a single run.

  • martin_pausewrite action

    Write a durable pause receipt for one canonical MartinLoop run. Provide exactly one selector: file, loopId, or latest; runsDir changes the run-store root, while reason and requestedBy add audit context. Use for a temporary hold before risky follow-up work. This records a request and does not kill a process; use martin_cancel to abandon work or martin_continue to resume.

  • martin_plan

    Read-only planning step that turns an objective into bounded scope, verifier proposal, policy pack, and risk recommendation. Use for authorized software changes before preflight/run. Do not use to mutate files or replace the verifier. Next: call martin_preflight with the chosen scope, budget, and verifier.

  • martin_pr_summary

    Generate a GitHub-ready pull-request title and Markdown body from a completed MartinLoop run dossier, including its verification grade and score.

  • martin_preflight

    Read-only validation of the exact run contract before execution or spend. Use after planning and before martin_run to check engine, verifier, path scope, and budget. Do not use as execution or completion proof. Next: call martin_run if allowed, otherwise resolve the reported blocker.

  • martin_review_pr

    Review a PR body or draft against the MartinLoop run dossier and evaluation evidence, and return a verdict with specific findings.

  • martin_runwrite action

    Execute a governed MartinLoop coding run after MCP workflow admission is satisfied. Use when the user has authorized implementation, bug fixing, tests, or refactoring and doctor/estimate/plan/preflight receipts match this task. Do not use for question-only diagnosis or when policy, budget, credentials, or scope still need consent. Next: read martin_dossier and verifier evidence.

  • martin_run_dossierwrite action

    Read the full structured execution dossier for one MartinLoop run, including attempts, events, artifacts, verification, integrity, cost, and related discovery surfaces. Provide exactly one selector: file, loopId, or latest; runsDir changes the run-store root. Use for comprehensive evidence review. Do not use for a compact state check; use martin_get_run, or use martin_dossier when formatted sharin

  • martin_status

    Read the current budget, cost, remaining limits, and stop pressure for one MartinLoop run. Provide exactly one selector: loopJson for an inline record, file for a saved record, loopId for a run-store ID, or latest for the newest run; runsDir only changes the run-store root. Use for a compact budget check. Do not use for full events or artifacts; use martin_get_run or martin_run_dossier instead.

  • martin_triage_runs

    Read-only prioritization of saved Martin runs that need attention. Use when the user says a prior attempt failed, asks what to fix next, or resumes an interrupted session. Do not use for a brand-new objective with no relevant run history. Next: inspect the selected run or dossier before retrying.

Public scan report

scanner v0.1.9 · 2026-09-20 · same rubric, same numbers if you re-run it

no findings
  • Code scan68 source files scanned25/25
  • –Live reliabilityno gateway calls yet and no remote to proben/a
  • –Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitylocal package, no credentials required12/15
  • Maintenancelast push 1 days ago15/15
  • Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
Overall 92/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

What the publisher says

From the Martin Loop repository's README, as published. We do not edit it. Read it on GitHub

MartinLoop

Your coding agent says it's done. MartinLoop makes it prove it.

One system to control, verify and understand coding-agent work.

MartinLoop gives AI coding agents budgets, stop conditions, rollback rules, and receipts.

Built from thousands of agent runs where the problem was not intelligence -- it was uncontrolled execution.

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

Install — 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.

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 execution-control system around coding agents. The coding agent still writes the code. MartinLoop governs the run and makes completion prove itself.

Use MartinLoop when a coding task needs one or more of these controls:

  • 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 control layer across Codex, Claude Code, and other coding-agent environments

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

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

Install directly

Runs npx -y @martinloop/mcp on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add martin-loop -- npx -y @martinloop/mcp
Add to Cursor

Martin Loop: common questions

Is Martin Loop MCP server safe?
Yes, by our scan: it is graded A (92/100). Read the Martin Loop safety report
How do I install Martin Loop?
It runs on your machine. Copy the Claude Code, Claude Desktop or Cursor config from the install section.
Does Martin Loop need an API key?
Not as far as the registry entry and our scan can tell: no credentials are declared or required.
Is Martin Loop maintained?
The last commit was 2 days ago (2026-09-19). The latest release is v0.6.1.

More from Keesan12 →