retro
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes session logs and repository steering files, creating an attack surface where malicious instructions embedded in historical data could influence the agent's behavior during the retrospective process.
- Ingestion points: The skill reads session logs,
package.json, CI workflows, and environment-steering files likeAGENTS.mdandCLAUDE.mdto identify candidates for improvement. - Boundary markers: There are no explicit instructions to use delimiters or to disregard embedded instructions within the logs or configuration files being analyzed.
- Capability inventory: The skill has the capability to read local files and suggest modifications to high-privilege instruction files (e.g.,
AGENTS.md,CODING_STANDARDS.md), which steer the behavior of the agent in subsequent sessions. - Sanitization: The skill does not implement sanitization or filtering of the ingested log content before it is processed by the agent to generate improvement candidates.
Audit Metadata