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Use when the user wants to batch-close resolved or outdated tracker items. The agent never closes tracker items itself. Not for individual closure or items still under active work.
$ npx -y skills add OutlineDriven/odin-claude-plugin --skill close-done --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/close-doneContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when the user wants to batch-close resolved or outdated tracker items. The agent never closes tracker items itself. Not for individual closure or items still under active work.
name: close-done description: 'Use when the user wants to batch-close resolved or outdated tracker items. The agent never closes tracker items itself. Not for individual closure or items still under active work.' disable-model-invocation: true
| Field | Bound contract | |---|---| | Trigger | User wants to batch-close resolved or outdated tracker items. | | Authority | Read-only. No file, VCS, credential, paid, published, deployed, or remote mutation. The agent fetches, classifies, and presents the batch; the user performs every remote closure. | | Side effect | A prepared batch list presented to the user. No tracker item is mutated by this skill. | | Done | A prepared batch list is presented to the user for manual closure, with each item classified as resolved or outdated and backed by evidence. |
An item qualifies for the batch only when it meets one of these evidence-backed criteria:
Items that do not meet either criterion are excluded. List each excluded item with the reason it failed the criteria.
1. Resolve the supplied references or filter against the tracker read-only and fetch each item's number, title, current state, last activity timestamp, and linked PRs or branches. Done when: every supplied item is resolved with its metadata or the failing item is named. 2. Classify each item as resolved or outdated using the criteria above. Drop items that are neither and list them as excluded with the reason. Done when: every item has a classification with its evidence or an exclusion reason. 3. Prepare the final batch list: every item proposed for close with its number, title, current state, classification (resolved or outdated), and the evidence supporting it. Done when: the full batch is listed and no mutation has occurred. 4. Present the batch to the user for manual closure. State that the agent will not close any item and that the user must perform the closure on the tracker. Include the optional resolution note for the user to attach. Done when: the batch is presented and the user has the closure instructions.
A batch report listing each item proposed for close (number, title, current state, classification, evidence), excluded items with reasons, and the optional resolution note for the user to attach. The report includes the tracker URL or instructions for the user to perform the closure manually.
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.