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 a coding-agent session store is added or its format drifts, to document its layout, roles, quirks. Not for recalling a session: use history-recall. Not for peer transfer: use history-sync.
$ npx -y skills add OutlineDriven/odin-claude-plugin --skill history-source-registry --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/history-source-registryContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when a coding-agent session store is added or its format drifts, to document its layout, roles, quirks. Not for recalling a session: use history-recall. Not for peer transfer: use history-sync.
name: history-source-registry description: 'Use when a coding-agent session store is added or its format drifts, to document its layout, roles, quirks. Not for recalling a session: use history-recall. Not for peer transfer: use history-sync.'
| Field | Bound contract | |---|---| | Trigger | Maintainer or user adds or verifies a coding-agent session store when its format drifts. | | Authority | Reversible local: writes only registry JSON, documentation pages, credential-free fixtures, and tests; rollback is version control. No remote mutation. | | Side effect | Updates repository-owned registry, credential-free fixture, and tests. | | Done | Observed discovery/layout/roles/time/quirks are documented; registry and page dates match; loader list cross-checks; a synthetic fixture passes; drift reports only the smallest redacted record, never a real transcript. |
1. **Discover the store layout.** Read the session store directory. Identify the file structure: transcript files, metadata files, index files, and nested directories. Record the glob patterns that match transcript-bearing files.
2. **Extract roles and time handling.** Parse one representative transcript file. Identify speaker roles (user, assistant, system, tool). Identify the timestamp format and timezone. Note quirks: missing timestamps, role encoding differences, non-standard field names, or embedded tool-call structures.
3. **Bound scope.** Do not read real transcripts beyond the minimum needed to extract layout, roles, time, and quirks. Use the smallest representative record. Never copy or expose real transcript content into documentation or fixtures.
4. **Create or update the registry entry.** Add or modify the store entry in the registry JSON. Include these fields: identifier, discovery glob, layout summary, roles, time format, known quirks, page date, and documentation page path. Ensure the page date matches the current date.
5. **Write or update the documentation page.** Create or update the per-source markdown page under the documentation root. Explain how to discover transcripts, the file layout, role encoding, timestamp format, and known quirks. Use a synthetic redacted example, never a real transcript.
6. **Create or update the conformance fixture.** Under the fixture root, create or update a minimal synthetic fixture that exercises the store discovery glob and loader path. The fixture must contain no real credentials or transcript content. It must be loadable by the test harness.
7. **Cross-check the loader list.** Read the loader test file. Verify every identifier in the registry has a corresponding test case. If a test case is missing, add one that loads the synthetic fixture and asserts the expected layout fields.
8. **Run the conformance test.** Execute the loader test. Confirm the new or updated fixture passes. If it fails, diagnose the fixture or loader, fix within bounded scope, and re-run. Do not widen scope beyond the failing assertion.
9. **Verify drift detection.** If re-verifying an existing store whose format drifted, confirm the registry entry, documentation page, and fixture all reflect the new format. The drift report names only the changed fields using the smallest redacted record.
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.