Menu
Generate optional advisory MAID run reviews outside maid-runner with cloud privacy disclosure.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Review an implementation produced from an approved MAID manifest. Confirms changed files stay within manifest scope, declared artifacts exist, validations pass, and the behavior matches the contract. Use after maid-implementer or after any MAID-backed code change before merge.
Implement code against an approved MAID manifest. Loads only the declared files, writes code to pass behavioral tests, validates with `maid validate --mode implementation`, and iterates until all checks pass. Use after a manifest is approved by maid-planner (or manually).
Plan a coding task as a machine-checkable MAID manifest instead of free-form markdown. Analyzes the project, asks clarifying questions, drafts or refines a manifest, and for implementation-ready plans adds behavioral tests, validates with `maid validate --mode behavioral`, and gets user approval. The promoted manifest becomes the implementation contract. Use at the start of every new feature, bug fix, or refactor.
Use when implementing maid-runner MAID draft manifests from manifests/drafts, promoting them into manifests, and validating implementation batches with review before handoff in Claude Code.
Review an implementation produced from an approved MAID manifest. Confirms changed files stay within manifest scope, declared artifacts exist, validations pass, and the behavior matches the contract. Use after maid-implementer or after any MAID-backed code change before merge.
Implement code against an approved MAID manifest. Loads only the declared files, writes code to pass behavioral tests, validates with `maid validate --mode implementation`, and iterates until all checks pass. Use after a manifest is approved by maid-planner (or manually).
| name | maid-run-review |
| description | Generate optional advisory MAID run reviews outside maid-runner with cloud privacy disclosure. |
Generate an optional LLM-assisted review of one completed MAID run while keeping maid-runner itself deterministic and AI-free. This skill is the only place in the run-review workflow that may call a model or touch provider credentials.
maid evaluate only for run evidence,
prompt request creation, review validation, and markdown rendering.--diff-file.maid evaluate options. Do not invent provider, model,
base URL, timeout, or API-key options for maid-runner.Show the deterministic run evidence first:
maid evaluate run <manifest>
Build the bounded review request:
maid evaluate prompt <manifest> --output .maid/run-review-request.json
If the operator wants diff evidence included, prepare a diff file explicitly, say how it was prepared, and pass it through the existing option:
maid evaluate prompt <manifest> --diff-file <diff-file> --output .maid/run-review-request.json
Generate a candidate review with an agent-available model.
Send only .maid/run-review-request.json to the model. If that uses a cloud
model, disclose CLOUD-PRIVACY first and name what run evidence leaves the
machine. The model output must be a JSON run review that cites only evidence
ids from the request.
Validate the candidate review:
maid evaluate validate .maid/run-review.json --request .maid/run-review-request.json
Treat any validation failure as fail closed. Do not hand-edit citations or evidence ids to force validity. Regenerate or correct the candidate review with the model from the bounded request, then validate again.
Render the validated advisory report:
maid evaluate render .maid/run-review.json --request .maid/run-review-request.json --output .maid/run-reviews/<slug>.md
Present the labeled advisory report to the human operator. Do not treat the report as a workflow gate, and do not give it to the implementing agent as mid-run instructions.
.maid/run-review-request.json, .maid/run-review.json, and
.maid/run-reviews/*.md are generated advisory artifacts. They are not
authoritative Outcome records and must not replace manifest outcome:
sections, validation commands, implementation review, or the MAID lifecycle.