PR Advocacy

Monitor PR status every 4 hours, promptly address feedback and CI issues, communicate clearly within 24 hours, and drive reviews to timely merge.

This skill will (per its instructions) run gh/git/npm commands, modify repository branches, and write a tracking file to a hard-coded home path — yet the registry metadata does not declare the required binaries, credentials, or config path. Before installing: 1) Verify you want an automated agent that can commit and push changes; prefer a bot account with limited repo scopes rather than your personal credentials. 2) Confirm GH CLI/git/npm availability and how authentication will be provided (GH_TOKEN or gh auth), and require the skill to declare those env vars. 3) Fix the hard-coded path: update it to a configurable, relative path or prompt for the correct location. 4) Consider disabling autonomous invocation or require explicit approval before any commit/push. 5) Test in a sandbox repository first. If the author cannot justify the undeclared dependencies and the hard-coded filesystem access, treat the skill as untrusted.

Type: OpenClaw Skill Name: pr-advocacy Version: 1.0.3 The skill is classified as suspicious due to its instructions for the AI agent to perform powerful actions that, while necessary for its stated purpose, introduce significant attack surfaces. Specifically, the `SKILL.md` instructs the agent to 'Commit changes to existing PR branch' and 'Maintain version history through git commits' for its tracking file, implying `git commit` and `git push` capabilities. It also provides `gh` shell command examples, indicating shell execution. These capabilities (modifying code in repositories, extensive file system interaction, and shell command execution) could be exploited via prompt injection or other vulnerabilities if the agent's input sanitization is not robust, even though the skill itself does not demonstrate explicit malicious intent like data exfiltration or backdoor installation.