.claude/skills/run-runbook/SKILL.md
Use when the user wants to launch a runbook, run an audit, review, or analysis task via Jetty, or says "run the X runbook". Also triggers on "launch runbook", "execute runbook", "run dedup-audit", "run security-reviewer", or any request to run a task from runbooks/.
npx skillsauth add ahrav/gossip-rs run-runbookInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Launch runbooks from runbooks/ via Jetty using run-runbook.sh.
runbook-params.json with any parameter overrides the user requested./run-runbook.sh <task-name>If runbook-params.json is missing, create it locally at repo root before
launching. The repo .gitignore excludes both runbook-params.json and
run-runbook.sh, so these runbook helper files may exist only in your local
checkout.
NEVER write secrets or tokens into runbook-params.json. The run-runbook.sh
script injects GITHUB_TOKEN automatically via gh auth token at launch time.
Do not add a GITHUB_TOKEN field to the JSON file — it is unnecessary and
risks leaking credentials.
{
"repository": "ahrav/gossip-rs",
"pr_number": "312",
"base_branch": "main",
"CARGO_HOME": "/Users/<you>/.cargo",
"timeout_sec": 3600
}
| Parameter | Default | Override for |
|-----------|---------|-------------|
| timeout_sec | 3600 (1 hr) | Long audits: 7200 (2 hr), 10800 (3 hr) |
| pr_number | varies | PR-specific runbooks (reviews, comment response) |
| base_branch | main | Auditing a different branch |
Use the Edit tool to update runbook-params.json before running the script.
Restore the original value immediately after launching if the override was
one-time — the runbook runs remotely and does not re-read the local file.
./run-runbook.sh <task-name>
The <task-name> matches the filename in runbooks/ without the .md extension.
if [ -d runbooks ] && ls runbooks/*.md >/dev/null 2>&1; then
ls runbooks/*.md | sed 's|runbooks/||;s|\.md$||'
else
echo "No local runbooks found. Ensure local runbook assets are provisioned."
fi
The script prints trajectory and workflow IDs on success:
Run launched successfully:
Task: <task-name>
Trajectory ID: <id>
Workflow ID: <id>
Report these to the user so they can track progress in Jetty.
| Mistake | Fix |
|---------|-----|
| Using RemoteTrigger API directly | Use run-runbook.sh — it handles param injection and Jetty API |
| Forgetting to update timeout for long tasks | User-requested timeout goes in runbook-params.json timeout_sec |
| Passing wrong task name | Name must match a file in runbooks/ without .md. The script lists available names on error. |
| Not reporting trajectory ID | Always show the trajectory and workflow IDs in your response |
| Not knowing which runbook to run | Use the guarded runbook-listing command above and ask the user to pick one |
| Writing GITHUB_TOKEN into runbook-params.json | Never. run-runbook.sh injects the token via gh auth token automatically. Writing tokens into files risks credential leaks. |
development
Deep first-principles code explanation that builds real understanding through phased walkthroughs with diagrams. Covers algorithms, data structures, memory layout, concurrency patterns, and performance tricks — especially for systems code in Rust. Use whenever the user asks to explain, walk through, break down, deep dive into, or understand code. Trigger on "how does this work", "what's happening here", "teach me about this", "why is it done this way", or when the user references a file with @ and wants to understand it. Proactively use when examining code involving lock-free algorithms, atomics/CAS, memory ordering,
development
Use when creating implementation-ready beads tasks that need testing strategy, optimal implementation approach, and documentation requirements baked in — composes /create-task with parallel enrichment agents that analyze the codebase and produce concrete test specifications, algorithm/data-structure guidance, and doc quality standards so implementing agents don't need to re-research
development
--- name: autoresearch description: Autonomous Goal-directed Iteration. Apply Karpathy's autoresearch principles to ANY task. Loops autonomously — modify, verify, keep/discard, repeat. Supports bounded iteration via Iterations: N inline config. version: 1.9.11 --- # Claude Autoresearch — Autonomous Goal-directed Iteration Inspired by [Karpathy's autoresearch](https://github.com/karpathy/autoresearch). Applies constraint-driven autonomous iteration to ANY work — not just ML research. **Core id
development
Use when implementing a new feature and assessing coverage gaps, during periodic test hygiene, when test suites feel bloated, or before merging code that changes coordination or hot paths. Two-phase assess-then-improve testing pipeline.