Render a dashboard of the current Claude Code session — message counts, token usage, cache read/write, dollar cost, context %, lines changed, cost-per-turn sparkline, and a tool-call histogram — parsed from the session transcript JSONL by a bundled bash script. One bash call, zero analysis, zero confirmation. Optionally pass a session id to inspect a different session. Use when the user asks for session stats, usage, token or cost breakdown, or a session dashboard.
The English summary is being prepared. Showing the original description.
See how to installReview the instructions the agent will receive before you install.
One bash call, then relay. The bundled script finds the transcript at
~/.claude/projects/*/<session-id>.jsonl (current id comes from
CLAUDE_CODE_SESSION_ID in the bash environment), aggregates it with jq,
and prints the dashboard. All numbers come from the script — compute nothing
yourself.
bash {base_directory}/dashboard.sh [session-id]
Print the script's stdout verbatim inside a ```text code fence. After the fence, add at most one sentence — or nothing.
Non-zero exit → print stderr verbatim and stop. No retries, no fallback parsing.
— for fields it can't find. Never "fill in" missing values by reading
the JSONL yourself.! bash {base_directory}/dashboard.sh --colorNo reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Retire a closed issue's working artifacts out of specs/ into the GitHub wiki. Use only when the user asks to archive a finished issue, empty out specs/ for a done task, or move an issue's requirements/architecture/tasks/review/QA docs to the wiki.
The English summary is being prepared. Showing the original description.
One-shot conventional commit; bundled scripts adaptively gather diff context, draft the message, and stage+commit — pass `ask` to confirm first or add a hint for the title. Use only when the user asks for a commit.
The English summary is being prepared. Showing the original description.
Executes a QA specification (specs/qa/QA-<N>-<slug>.md) written by /plan-qa: drives every step with the plan's drivers, verifies each Expected line mechanically ([assert]) or by evidence-backed judgment against the plan's written criterion ([judge]), pauses at named operator handoffs, and writes the results artifact QA-RESULTS-<N>-<slug>.md. Use only when the user asks to run or execute a QA plan — never trigger automatically.
The English summary is being prepared. Showing the original description.
After an issue's PR has squash-merged on GitHub and its issue has closed: fast-forward the default branch, remove the issue-numbered nested worktree (.worktrees/<issue#>), and delete the local feature branch. Use when the user says to finish, tear down, or clean up a lane or worktree for a merged issue.
The English summary is being prepared. Showing the original description.
Phase 3 of 5 — slices the ARCH doc into verification-ready task specs (tdd/test-after/ui/checklist), emitted as a separate TASKS-<N>-<slug>.md file alongside ARCH for the implement skill. Use only when the user asks to run Phase 3 or generate tasks from an ARCH doc — never trigger automatically.
The English summary is being prepared. Showing the original description.
Phase 4 of 5 — implements tasks from TASKS-<N>-<slug>.md (with ARCH-<N>-<slug>.md as architecture-only context) using mode-appropriate verification (tdd, test-after, ui, checklist); pass 'auto' to run without stepping. Use only when the user asks to run Phase 4 or implement tasks from an ARCH doc — never trigger automatically from a coding request.
The English summary is being prepared. Showing the original description.