Triage

Move issues on the project issue tracker through a small state machine of triage roles.

If this repo treats external pull requests as a request surface (see the issue-tracker config), triage covers them too: a PR is an issue with attached code, using the same roles, same states, and same machine, with a few deltas marked "for a PR" below. Resolve a bare #42 to an issue or PR per the tracker config.

Every comment or issue posted to the issue tracker during triage must start with this disclaimer:

> *This was generated by AI during triage.*

Reference docs

• AGENT-BRIEF.md: how to write durable agent briefs
• OUT-OF-SCOPE.md: how the .out-of-scope/ knowledge base works

Roles

Two category roles:

• bug: something is broken
• enhancement: new feature or improvement

Five state roles:

• needs-triage: maintainer needs to evaluate
• needs-info: waiting on reporter for more information
• ready-for-agent: fully specified, ready for an AFK agent
• ready-for-human: needs human implementation
• wontfix: will not be actioned

For a PR, the same states read against the attached code: ready-for-agent means a brief is attached and an agent should take the next step on the diff; ready-for-human means it's ready for a human to merge.

Every triaged issue should carry exactly one category role and one state role. If state roles conflict, flag it and ask the maintainer before doing anything else.

These are canonical role names. The actual label strings used in the issue tracker may differ. The mapping is in docs/agents/triage-labels.md, and the tracker operations are in docs/agents/issue-tracker.md. Read both before proceeding. If either is missing or unreadable, stop and tell the user to run /setup-supermatt-skills in Claude linked installs or Cursor, /supermatt:setup-supermatt-skills from the Claude plugin, $setup-supermatt-skills in Codex, or /skill:setup-supermatt-skills in OMP. Resume only after both required docs exist and have been read.

State transitions: an unlabeled issue normally goes to needs-triage first; from there it moves to needs-info, ready-for-agent, ready-for-human, or wontfix. needs-info returns to needs-triage once the reporter replies. The maintainer can override at any time; flag transitions that look unusual and ask before proceeding.

Invocation

The maintainer invokes /triage in Claude linked installs or Cursor, /supermatt:triage from the Claude plugin, $triage in Codex, or /skill:triage in OMP, and describes what they want in natural language. Interpret the request and act. Examples:

• "Show me anything that needs my attention"
• "Let's look at #42" (issue or PR)
• "Move #42 to ready-for-agent"
• "What's ready for agents to pick up?"

When the maintainer explicitly delegates decisions for this run ("I am away", "decide for me"), record that request and the issue or queue scope. Under the delegated triage contract, recommendation, selection, and outcome-confirmation waits within that scope use your recommended option instead: record each choice and reason in the brief or comment, then continue. This includes follow-up issues only within the delegated scope. Select grilling's delegated mode when needed and pass that request and scope. Missing facts, out-of-scope decisions, and separate approval requirements remain unresolved; record the blocker and use needs-info or ready-for-human as appropriate, never ready-for-agent. Without explicit delegation, keep the interactive waits.

Show what needs attention

Use the configured listing operations, consuming every page, then obtain the same full fields as Gather context for every discovered issue or PR before classifying it. Preserve the reporter identity, creation timestamp, and every comment's author, body, and timestamps; a title/body-only projection is insufficient. Use the configured operations for custom trackers, supplementing incomplete list results with full reads. A failed or incomplete listing is not an empty bucket. If required metadata remains unavailable, report the affected items and the coverage gap rather than assigning invented values.

Present three buckets, oldest creation timestamp first within each bucket across issues and PRs (ticket number breaks ties):

1. Unlabeled: carries no triage role.
2. needs-triage: evaluation in progress.
3. needs-info with reporter activity since the last triage notes: needs re-evaluation.

For the third bucket, find the latest ## Triage Notes entry and compare comment creation timestamps: only a later comment by the original reporter counts. A newer maintainer comment or general item update does not. Missing reporter identity, timestamps, comments, or a triage-note baseline makes this classification unknown; report it separately rather than treating it as no reporter activity.

When PRs are in scope, include external PRs in these buckets and tag each line [PR] or [issue]. Discovery surfaces only external PRs (the tracker config defines who counts as external), so a collaborator's in-flight PR is not triage work. This filter is discovery-only; an explicitly named PR is always triaged regardless of author.

Leave out issues that other skills produced already triaged, recognisable by their templates: specs from /to-spec (a ## Problem Statement and ## User Stories), tickets from /to-tickets (a ## What to build), and anything labelled wayfinder:*.

Show counts and a one-line summary per item. Let the maintainer pick.

Triage a specific issue or PR

1. Gather context. Read docs/agents/issue-tracker.md first, in its own call. Then read the full issue or PR (body, comments, labels, author, dates; for a PR, the diff too) with that file's Read an issue command (Read a PR and the diff command for a PR). For GitHub JSON reads, request title,state,body,comments,labels,author,createdAt for either surface and remove any --jq projection so those values and comment metadata remain accessible. For other trackers, retain the equivalent full fields and all comment pages. Do not build a read command from memory: flag combinations that look right fail, such as gh issue view --comments --json. Parse any prior triage notes so you don't re-ask resolved questions. Explore the codebase using the project's domain glossary, respecting ADRs in the area. Run two checks against the codebase: (a) redundancy: search for an existing implementation of the requested behavior by domain concept (not just the request's wording), and report where you looked. If found, it's an already-implemented wontfix (step 5). (b) prior rejection: read .out-of-scope/*.md and surface any that resembles this request. Run both checks in a sub-agent that returns only what it found and where it looked.
2. Recommend. Tell the maintainer your category and state recommendation with reasoning, plus a brief codebase summary relevant to the request (including whether it's already implemented). Wait for direction.
3. Verify the claim. Before any grilling, check that the claim holds up. For a bug, reproduce it from the reporter's steps. For a PR, confirm the diff does what it claims: check it out, run the relevant tests or commands. Report what happened: confirmed (with code path), failed, or insufficient detail (a strong needs-info signal). A confirmed verification makes a much stronger agent brief. When verification disproves a statement in the issue or PR body, correct it where later readers look first: edit the body when the maintainer or an agent wrote it, and otherwise open the brief with the correction. Later readers, the PR writer too, read the body before the brief.
4. Grill (if needed). If the request needs fleshing out, load and follow the grilling and domain-modeling skills, each through the host's skill mechanism. Use grilling's interactive mode unless this run has the explicit delegation recorded above; then pass it that delegation and scope for delegated mode. Sharpen domain terms and update GLOSSARY.md and ADRs inline as evidenced decisions land. Unresolved facts or decisions stay visible in the outcome and do not become implementation-ready requirements.
5. Apply the outcome:
   • ready-for-agent: agree with the maintainer the seams the tests go through, then post an agent brief comment that records them (AGENT-BRIEF.md).
   • ready-for-human: same structure as an agent brief, but note why it can't be delegated (judgment calls, external access, design decisions, manual testing).
   • needs-info: post triage notes (template below).
   • For wontfix, close the issue, with the comment depending on why:
     • Already implemented: the change already exists in the codebase. Point to where it lives; do not write to .out-of-scope/ (that KB is for rejected requests, not built ones).
     • Rejected (bug): give a polite explanation, then close.
     • Rejected (enhancement): write to .out-of-scope/, link to it from a comment, then close (OUT-OF-SCOPE.md).
   • needs-triage: apply the role. Optional comment if there's partial progress.

Quick state override

If the maintainer says "move #42 to ready-for-agent", trust them and apply the role directly. Confirm what you're about to do (role changes, comment, close), then act. Skip grilling. Moving to ready-for-agent still needs an agent brief: ask the maintainer for the seams it records, then write it.

Needs-info template

> *This was generated by AI during triage.*

## Triage Notes

**What we've established so far:**

- point 1
- point 2

**What we still need from you (@reporter):**

- question 1
- question 2

Capture everything resolved during grilling under "established so far" so the work isn't lost. Questions must be specific and actionable, not "please provide more info".

Resuming a previous session

If prior triage notes exist on the issue or PR, read them, check whether the reporter has answered any outstanding questions, and present an updated picture before continuing. Don't re-ask resolved questions.