retro
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes session logs, which may contain untrusted data from prior agent interactions.\n
- Ingestion points: Session logs and repository files such as
package.jsonand CI configurations (SKILL.md).\n - Boundary markers: No instructions for delimiters or warnings to ignore embedded content within the processed data are provided.\n
- Capability inventory: The skill allows the agent to suggest and potentially implement modifications to repository documentation (
CODING_STANDARDS.md,CLAUDE.md), linter configurations, and CI workflows.\n - Sanitization: No sanitization or validation of ingested log data is specified.\n- [SAFE]: No malicious code, network exfiltration, or unauthorized credential access patterns were found in the skill's instructions or configuration. The identified risk is a standard surface for analysis-based skills.
Audit Metadata