ops
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill performs automatic detection of project context by scanning local files for "signals," which exposes the agent to untrusted data that could contain malicious instructions.\n
- Ingestion points: Local project files are analyzed to determine the appropriate monitoring templates (SKILL.md).\n
- Boundary markers: Absent. There are no delimiters or specific instructions provided to ensure that instructions within the scanned project files are ignored by the agent.\n
- Capability inventory: The skill is granted access to the
Read,Grep,Glob, andBashtools, providing significant access to the local environment (SKILL.md).\n - Sanitization: Absent. No sanitization or validation logic is defined for the content read from project files before it influences the agent's behavior.\n- [COMMAND_EXECUTION]: The skill uses the
Bashtool to run an internal script,lib/context-detector.sh, to automate the analysis of the user's project environment (SKILL.md).
Audit Metadata