| name | project-status |
| description | On-demand project status report covering recent work, current state, and drift between code, docs, and specs. |
| license | MIT |
| compatibility | ["claude-code"] |
| metadata | {"author":"cdimonaco","version":1,"generatedBy":"human"} |
Project Status Report
Generate a concise status report for the tokenpile project. Covers what was done recently, what is active now, and whether code, docs, and specs appear to be in sync.
Steps
-
Run these commands in parallel:
git log --oneline -15
git status --short
git branch --show-current
openspec list --json 2>/dev/null || echo '{"changes":[]}'
gh run list --limit 3 --json status,name,conclusion,createdAt 2>/dev/null || echo '[]'
-
Read in parallel:
README.md — features list and commands section (first 120 lines is enough)openspec/specs/ — list the directory to see which capabilities are documented
-
Analyze for drift. Check each of these:
- Do recent commits (feat: or fix:) touch packages not mentioned in README features?
- Are there active OpenSpec changes with no corresponding recent commits?
- Are there OpenSpec specs for capabilities that appear to be unimplemented (no matching package in
internal/)?
-
Produce the report in this exact format:
Project Status — tokenpile
Recent work
<list the last 15 commits grouped by type: feat, fix, test, chore, docs, refactor>
Current state
- Branch:
- Uncommitted: <"clean" or list of files>
- CI: <last run name + conclusion, or "unknown">
- Active OpenSpec changes: <list change names or "none">
Documented capabilities
<one line per file found in openspec/specs/, e.g. "usage-tracking — core entry/session model">
Drift signals
<bullet list of concrete inconsistencies found, or "none detected">
Rules
- Flag drift only when there is clear evidence. Do not speculate.
- Keep each section tight — one line per item where possible.
- If a tool is unavailable (openspec, gh), note it in the relevant section and continue.
- Do not implement or change anything. This skill is read-only.
