agent-watchdog
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill functions as a code review and session auditing utility. Its operations are intended to verify the integrity of work performed by other agents, utilizing standard environment tools for file reading and repository inspection.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of processing data from external, non-authoritative sources.
- Ingestion points: The skill ingests session IDs, transcript paths, thread URLs (GitHub, Slack), repository logs, and pasted summaries (SKILL.md).
- Boundary markers: No explicit delimiters or instructions are provided to the agent to treat audited content as untrusted or to ignore embedded instructions.
- Capability inventory: The agent can read files, inspect git diffs, execute arbitrary commands for validation, and modify code files to repair gaps (SKILL.md).
- Sanitization: No input validation or content sanitization is mentioned in the skill instructions.
- [COMMAND_EXECUTION]: The skill requires the agent to execute shell commands and test suites to verify evidence and validate fixes. This capability is necessary for the skill's primary objective of auditing and repairing code.
Audit Metadata