secrets-rotation
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from chat messages, logs, and commits to identify exposed secrets. This represents an attack surface for indirect prompt injection.\n
- Ingestion points: Operator messages, project logs, git commits, and transcripts.\n
- Boundary markers: Absent.\n
- Capability inventory:
Read,Bash, andGreptools.\n - Sanitization: The instructions advise against repeating the secret value, focusing instead on metadata (kind and location), which partially mitigates the risk.\n- [COMMAND_EXECUTION]: The skill directs the agent to execute shell commands using the
Bashtool to create tasks in a tracking system (bd create). User-influenced metadata such as the 'kind' or 'location' of a secret is interpolated into these commands, which could lead to command injection if the agent does not properly escape the input.
Audit Metadata