agent-environment-retr…
Use when a completed session needs an agent-environment retrospective. Not for an engineering retrospective from telemetry: use engineering-retrospective.
Use when evaluating whether a bug issue has sufficient detail and identifying missing reporter information. Not for non-bug issues or backlog triage: use github-backlog-triage.
$ npx -y skills add OutlineDriven/odin-claude-plugin --skill github-bug-report-triage --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/github-bug-report-triageContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when evaluating whether a bug issue has sufficient detail and identifying missing reporter information. Not for non-bug issues or backlog triage: use github-backlog-triage.
name: github-bug-report-triage description: 'Use when evaluating whether a bug issue has sufficient detail and identifying missing reporter information. Not for non-bug issues or backlog triage: use github-backlog-triage.' disable-model-invocation: true
| Field | Bound contract | |---|---| | Trigger | Evaluating whether a bug issue has sufficient detail and identifying missing information from the reporter. | | Authority | Remote: drafts one missing-information comment for a GitHub issue and previews it with the target issue; the human posts it or applies any label. Requires explicit human invocation; the model never posts, labels, closes, or otherwise mutates the issue or repository on its own. | | Side effect | A comment on the GitHub issue requesting missing information, or a confirmation that the issue is actionable. Posting is a remote mutation performed or explicitly authorized by the human; the model never posts autonomously. | | Done | Ready issues are confirmed actionable; missing-info issues receive specific, constructive feedback listing exactly what is needed. |
A GitHub bug issue (number or URL) in the target repository. Required.
The project bug-report issue template, if one exists. Optional; located by the procedure.
1. Bound scope to a single bug issue. Do not edit, label, close, or post anything until the human authorizes. Done when: scope is bounded to a single bug issue, and no edit, label, close, or post has been made. 2. Read the issue title, body, and existing comments. Done when: the issue title, body, and existing comments are read and recorded. 3. If the issue is a feature request or a question, stop: do not evaluate it as a bug report. Done when: the issue is confirmed to be a bug, or the run stops with a stated classification as feature request or question. 4. Locate the project bug-report template in this order: `.github/ISSUE_TEMPLATE/bug_report.md`, `.github/ISSUE_TEMPLATE/bug_report.yml`, `.github/ISSUE_TEMPLATE/*.md`, `.github/ISSUE_TEMPLATE.md`, `ISSUE_TEMPLATE.md`. If none exists, evaluate the issue against this bug-report field set: Summary (what is broken), Expected behavior, Actual behavior (including any error messages), Steps to reproduce or a reproduction link / minimal code example, and Environment (OS, runtime, version or commit, install method). Done when: the bug-report template is located at one of the searched paths, or its absence is confirmed and the default field set is applied. 5. Judge actionability against the minimum information needed to investigate, not against every template field being filled: a clear problem description plus a reproduction path (steps, link, or minimal code) or enough context to debug is sufficient. Done when: actionability is judged against the minimum information needed to investigate (problem description plus reproduction path or debug context), and the judgment is recorded with its reasoning. 6. Classify the issue. Ready: clear problem description, a reproduction path or sufficient debug context, and actually a bug. Needs-info: vague description, missing reproduction and no code example, unclear expected behavior, or missing critical context such as an error message. Done when: the issue is classified as Ready or Needs-info, and the classification names the specific evidence or gap that determined it. 7. If Ready, confirm the issue is actionable. If Needs-info, list each specific missing item constructively. Done when: a Ready issue is confirmed actionable, or a Needs-info issue has each specific missing item listed constructively. 8. Draft the comment text. Preview the comment and the target issue to the human. The human posts the comment or applies any label; the model does not post, label, or close autonomously. Done when: the comment text is drafted and previewed to the human with the target issue, and no comment is posted, no label applied, and no issue closed by the model.
A determination of Ready or Needs-info with reasoning, plus a drafted comment. For Needs-info the comment lists each specific missing item. No issue is mutated until the human posts the comment.
Formerly the ODIN Claude Plugin. The repository URL is unchanged. Outline-Driven Development, nicknamed ODIN, is a highly opinionated code-agent skill library: principles-first engineering, surgical editing, and workflow automation, published as installable
Repo: OutlineDriven/odin-claude-plugin
Use when a completed session needs an agent-environment retrospective. Not for an engineering retrospective from telemetry: use engineering-retrospective.
Use when a repo needs agent setup, AGENTS.md added or made lean, CLAUDE.md audited, or agent instructions scored or pruned. Not for remote, credential,…
Use when a human explicitly asks for a full repository agent-compatibility pass returning a scored report with prioritized fixes. Not for tasks that require…
Use when setting up a project, auditing agent command permissions, or asking which read-only bash commands and domains to allow. Not for remote, credential,…
Use when asked to build or review a CLI intended for coding agents and return flag-driven, pipeline-safe, idempotent design advice. Not for running or…
Use when the user asks to make the skills framework work in a new harness, IDE, or CLI. Not for remote, credential, publish, deploy, or irreversible changes.