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 an artifact or skill has just changed and is about to be called done, committed, or handed off. Not for remote, credential, publish, deploy, or irreversible changes.
$ npx -y skills add OutlineDriven/odin-claude-plugin --skill post-change-check-gate --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/post-change-check-gateContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when an artifact or skill has just changed and is about to be called done, committed, or handed off. Not for remote, credential, publish, deploy, or irreversible changes.
name: post-change-check-gate description: 'Use when an artifact or skill has just changed and is about to be called done, committed, or handed off. Not for remote, credential, publish, deploy, or irreversible changes.'
| Field | Bound contract | |---|---| | Trigger | Right after creating or changing an artifact/skill, before calling it done, committing, or handing off | | Authority | Reversible local: writes only the named artifact; rollback is revert to pre-check state. No remote mutation. | | Side effect | Applies findings to the artifact, or defers each with a stated reason; never touches git | | Done | Relevant checks actually ran (not eyeballed); artifact changed or every skipped check has a stated reason |
1. **Run checks inline.** Do not accept "looks fine" or "eyeballed" as proof. For each applicable hygiene concern (lint, type, test, format, link, security scan, or domain-specific validation), invoke the check and capture its concrete pass/fail/skip output. Done when: every applicable check is invoked with concrete output captured. 2. **Log every result.** Record pass, fail, or skip with the specific check name and a one-line reason for skip. Done when: every check result is logged with name and outcome. 3. **Apply fixes only to the artifact.** When a check fails, edit the artifact to fix the finding. Do not edit unrelated files. Done when: every failed check's finding is fixed in the artifact or deferred with a reason. 4. **Record skip reasons.** If a check is skipped, write the reason explicitly into the check log before proceeding. Done when: every skipped check has an explicit reason in the log. 5. **Verify the check set passes.** Re-run checks until all relevant checks pass or every skip is documented. If a check cannot be made to pass, stop and leave the artifact in a documented deferred state; do not commit or hand off a dirty artifact. Done when: all relevant checks pass or every skip is documented and the artifact is in a clean or documented-deferred state.
The artifact with all check findings applied, or the check log recording every skip with its reason, plus a concrete check-run report naming artifact path, checks invoked, results per check, and any deferred items with stated reasons.
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.