agentleFS
runxhqrunxhq/runx/skills/issue-triage/SKILL.md

issue-triage

Skill97 starsChanged 33 days agoApache-2.0View on GitHub

Read, analyze, and draft high-signal GitHub issue responses from bounded provider evidence without silently mutating the repository.

63 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use meeting-notes-and-actions, pr-review-ci-fix and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • ComposioHQSkill

    meeting-notes-and-actions

    Turn meeting transcripts or rough notes into crisp summaries with decisions, risks, and owner-tagged action items; use for Zoom/Meet/Teams transcripts, call notes, or long meeting chats to generate share-ready outputs.

    Also in 4 projects that use this

  • ComposioHQSkill

    pr-review-ci-fix

    Automated PR review and CI auto-fix for GitHub and GitLab using the Composio CLI. Pulls diffs, fetches failing job logs, posts review comments, and loops fix commits until checks go green.

    Also in 4 projects that use this

  • FlorianBruniauxSkill

    tdd-rust

    Also in 3 projects that use this

  • GitHubSkill

    phoenix-cli

    Debug LLM applications using the Phoenix CLI. Fetch traces, analyze errors, structure trace review with open coding and axial coding, inspect datasets, review experiments, query annotation configs, and use the GraphQL API. Use whenever the user is analyzing traces or spans, investigating LLM/agent failures, deciding what to do after instrumenting an app, building failure taxonomies, choosing what evals to write, or asking "what's going wrong", "what kinds of mistakes", or "where do I focus" — even without naming a technique.

    Also in 3 projects that use this

  • petestewartSkill

    datadog-logs

    Query Datadog logs via the Logs Search API v2. Use when the user asks to "check the logs", "search logs", "find errors in Datadog", "look up logs for", "show me production logs", "check for errors", "search Datadog", or any request involving log search, error investigation, or production debugging that requires Datadog log data.

    Also in 3 projects that use this

  • +7 more from ComposioHQ

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 1 is distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 51 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use the default runner when the operator has a repository and issue number and wants current provider-grounded evidence.

    line 25
  2. Very specific:

    Do not use the analysis runners to close, label, assign, comment on, or promise work in a repository.

    line 32
  3. Very specific:

    Every candidate and response is bound through Runx's native data digest to the exact issue snapshot or bounded snapshot set.

    line 38
  4. Distinctive:

    Do not turn maintainer context into provider evidence.

    line 70
  5. Very specific:

    Keep mutation out of the analysis lanes.

    line 73

“Common” is measured: kept by at least 6 in 10 of the 51 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Issue Triage

Turn a noisy issue queue into one evidence-backed next action. The skill helps a maintainer decide which thread deserves attention and how to respond without silently mutating a repository or inventing state that is not present in the thread snapshot.

Discovery and response are separate jobs. The default provider-respond runner reads one issue through the GitHub instance selected here, the signed-in gh session or a stored token, binds the returned snapshot and readback into the receipt, and turns it into a concise profile, recommended posture, draft reply, and follow-up plan. discover and respond remain explicit supplied-evidence lanes for offline queues and replay. Every analysis lane stops at a draft. Select post explicitly to consume that exact packet under a scoped comment grant and require provider readback without repeating triage.

When to use it

Use the default runner when the operator has a repository and issue number and wants current provider-grounded evidence. Use discover for a supplied issue queue and respond when an exact snapshot is already available or a receipt is being replayed. Use issue-intake when the material must first become an engineering change request, and issue-to-pr only after explicit promotion to implementation.

Do not use the analysis runners to close, label, assign, comment on, or promise work in a repository. They prepare the maintainer decision and draft only; the explicit post runner performs only the exact accepted comment.

Evidence and provider boundary

Every candidate and response is bound through Runx's native data digest to the exact issue snapshot or bounded snapshot set. provider-respond obtains that snapshot through github/issue.read with repo.read, and the provider operation remains in the sealed run alongside the draft. post comments through github/comments.create, which completes only when the read-back comment matches the exact accepted body. Supplied snapshots are labelled as such rather than promoted to provider readback.

Ground the assessment in the thread, repository facts, receipts, and supplied maintainer context. Do not infer contributor intent, reproduce hostile language unnecessarily, or claim a fix, release, or investigation exists unless the snapshot proves it.

Inputs and result

discover accepts a bounded issue_snapshots set, a selection query, and optional maintainer constraints. It returns a ranked triage queue whose ids and rationales bind to the admitted snapshot index.

provider-respond accepts owner/name, an issue number, and optional objective and maintainer context. respond accepts the same analysis context with one already obtained issue_snapshot. Both return the issue profile, response strategy, unsent response draft, and concrete follow-up actions with delivery_status: not_sent; the provider status distinguishes live readback from supplied evidence.

Stop conditions

  • Return needs_more_evidence or needs_human when a thread is ambiguous, hostile, unsafe, underspecified, or outside declared maintainer posture.
  • Reject issue ids, repository state, labels, commitments, or completed work not present in the admitted snapshot.
  • Do not turn maintainer context into provider evidence.
  • Refuse when no GitHub instance is selected, or the selected one cannot read the issue; the skill never handles a raw token.
  • Keep mutation out of the analysis lanes. An accepted draft moves explicitly to post, which requires scoped GitHub comment authority and provider readback.

Example

A queue contains a reproducible regression, a broad feature request, and a stale question. Discovery can prioritize the regression and explain why. Response can draft a concise request for the missing version detail or describe a verified workaround. It cannot say “fixed in the next release” unless the snapshot or other admitted evidence establishes that fact. Posting requires the separate post invocation and its comment-scoped grant.

Agent task contracts

issue-triage-discover

Rank only issues in the supplied index. Return bounded candidates and the selection rationale an operator needs to review the ranking. Never invent issue ids, provider state, promises, or completed work.

issue-triage-respond

Draft one helpful maintainer response grounded only in the admitted issue evidence. Preserve repository, issue number, title, and state exactly. Do not claim work is complete unless the snapshot proves it. The draft is not sent.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.