debug
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to capture and trace failures by examining error messages, stack traces, and logs. This behavior creates an ingestion point for untrusted data that could contain malicious instructions designed to manipulate the agent's analysis or subsequent actions.
- Ingestion points: Logs, error messages, and stack traces processed in the 'Trace the failure' step of SKILL.md.
- Boundary markers: The instructions do not specify delimiters or security warnings to isolate log content from instructional context.
- Capability inventory: The skill facilitates command execution for repository history and troubleshooting, such as
git logandgit bisect. - Sanitization: No specific sanitization or filtering logic is provided for the data ingested from logs or error outputs.
Audit Metadata