personalize-repo
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple sources, including existing repository documentation (AGENTS.md, ARCHITECTURE.md, etc.), the local REPO_PROFILE.json, and user-provided answers during the personalization interview. This data influences the agent's proposals for system governance and architecture updates.
- Ingestion points: The skill reads various markdown files and JSON profiles from the target repository root and captures natural language input from the user during the 'Vibe check interview'.
- Boundary markers: The instructions do not define specific delimiters or guardrail prompts to prevent the agent from following instructions potentially embedded within the ingested project files.
- Capability inventory: The skill has permissions to read and write files within the repository, including updating governance docs and generating the canonical REPO_PROFILE.json.
- Sanitization: There is no evidence of sanitization or filtering for the content ingested from the repository docs before it is used to suggest architectural or behavioral changes.
Audit Metadata