kf-g-command-do-pr-rev-comment
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data from GitHub PR review comments, which serves as an injection vector for malicious instructions.
- Ingestion points: SKILL.md (Collecting and categorization of unresolved review threads).
- Boundary markers: The skill does not define specific delimiters or instructions to treat the comment content as data only.
- Capability inventory: The agent has the ability to execute git commit, git push, and a local bot script for GitHub interactions (SKILL.md, references/group-commit-reply-sequence.md).
- Sanitization: No sanitization or validation of the review comment content is specified before it is processed by the agent.
- [COMMAND_EXECUTION]: The skill explicitly instructs the agent to execute shell commands, including git operations and a local Node.js bot helper script located at ~/.agents/credentials/github/scripts/github-agent-reply.mjs (SKILL.md).
Audit Metadata