About extract-todos
A skill for finding TODO, FIXME, HACK, and XXX markers across a codebase. It returns a prioritized inventory grouped by file, with counts, context, and rationale for suggested priorities and stale candidates.
Key features
- 01Finds TODO-like markers
- 02Groups results by file
- 03Provides counts and context
- 04Suggests priority and stale candidates
Use cases
- 01Inventorying deferred work in a codebase
- 02Triage of TODO-like comments
- 03Identifying potentially stale tasks
Install
Copy the command and run it in a terminal.
$
npx skills add https:/ / github.com/ cowork-os/ cowork-os --skill extract-todosAfter installing, type/extract-todosin Claude Code to confirm it loaded.
Details
From the project READMEExtract TODOs
Purpose
Find all TODO/FIXME comments in codebase
Routing
- Use when: Use when the user wants TODO/FIXME/HACK/XXX inventorying, triage, or stale-task cleanup.
- Do not use when: Don't use for implementation requests or when a code review already includes a TODO pass.
- Outputs: Prioritized TODO inventory with counts, file grouping, and stale candidates.
- Success criteria: Findings list all TODO-like markers with location, context, and suggested priority rationale.
Trigger Examples
Positive
- Use the extract-todos skill for this request.
- Help me with extract todos.
- Use when the user wants TODO/FIXME/HACK/XXX inventorying, triage, or stale-task cleanup.
- Extract TODOs: provide an actionable result.
Negative
- Don't use for implementation requests or when a code review already includes a TODO pass.
- Do not use extract-todos for unrelated requests.
- This request is outside extract todos scope.
- This is conceptual discussion only; no tool workflow is needed.
Parameters
| Name | Type | Required | Description |
|---|---|---|---|
| path | string | Yes | Path to search |
Runtime Prompt
- Current runtime prompt length: 517 characters.
- Runtime prompt is defined directly in
../extract-todos.json.
Comments
Markdown supported