runbook-generator

Generate operational runbooks from project files. Scans Dockerfiles, docker-compose.yml, systemd units, Makefiles, package.json, and config files to produce...

This skill appears to do what it says: it scans project infrastructure files and produces runbooks using only Python stdlib. Before running or giving an agent access to it: 1) Review scripts/generate_runbook.py locally to confirm behavior (no network calls or subprocesses were seen in the visible code). 2) Run it on a copy of your repo or in an isolated environment, especially if the project contains production secrets — the script reads .env and systemd unit files (it attempts to mask .env values but still reads them). 3) If you plan to allow autonomous agent invocation, consider restricting which directories the agent can point it to and avoid exposing host-level/system directories. 4) If you need stronger guarantees, ask the owner for source provenance or audit the rest of the file (the provided file was truncated in the review text); absence of detected network calls in the shown code is reassuring, but verify the full file before trusting it with sensitive projects.

Type: OpenClaw Skill Name: ops-runbook-generator Version: 1.0.0 The skill is a legitimate utility for generating operational documentation by scanning project configuration files such as Dockerfiles, systemd units, and Makefiles. The core logic in `scripts/generate_runbook.py` uses only the Python standard library, performs no network requests, and executes no shell commands. Notably, the script includes privacy-conscious logic to mask sensitive values when scanning `.env` files, and no evidence of prompt injection or malicious intent was found.