Execute a governed Martin Loop run on a coding task and return the run summary, spend, artifact rollup, and verification state. This hard-blocks until martin_doctor, martin_estimate, martin_plan, and martin_preflight receipts exist for the same task.

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

Return the current budget and cost state of a Martin loop record.

Read-only environment and run-store diagnostics for the Martin MCP server. This is the expected first call before governed work begins.

Read-only planning step that turns an objective into a scoped implementation plan, verifier proposal, policy pack, and risk recommendation. Use before preflight and before any real coding run.

Read-only validation of a planned Martin run before any execution or spend. This is the last required step before martin_run.

Read recent Martin loop events, ledger entries, and operator control receipts for live observability.

Record a durable pause request for a Martin run so humans and runtimes can see that execution should pause before risky follow-up work.

Record a durable cancellation request for a Martin run. This writes a control receipt; it does not silently kill a process without evidence.

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

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

Prioritize Martin runs that need operator or agent attention based on verification, lifecycle, and budget pressure.

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

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

Load verification evidence for a Martin run from stored loop events and ledger entries.

Return the full governed execution dossier for one Martin run, including attempts, events, artifacts, and related discovery surfaces.

Alias for martin_run_dossier with support for JSON, Markdown, or GitHub PR formatting. Use after martin_run to understand what happened and whether the result is actually safe to trust.

Grade a Martin run for task completion, verifier health, diff discipline, risk, and reviewability.

Generate a PR title and body with a MartinLoop dossier block for a completed run.

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

Review a PR or PR draft against the Martin dossier and evaluation evidence.

Estimate the cost, recommended route, and Pre Work Burn for an objective without spending anything. Use before martin_run to understand what a task will cost.