ptlam-git
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is subject to indirect prompt injection via repository data and local configuration files.
- Ingestion points: Git repository state (HEAD, refs, objects, diffs) and the project-local CONTEXT.md file located at .ptlam-skills/skills/engineering/ptlam-git/CONTEXT.md as described in references/project-git-context.md.
- Boundary markers: No explicit delimiters or 'ignore' instructions for untrusted data are defined for repository content.
- Capability inventory: Execution of Git commands (inspect, mutate, push) and file system writes for maintaining project context files.
- Sanitization: No explicit content filtering or sanitization is described, although the skill emphasizes a 'proof envelope' and 'principles' for verification.
Audit Metadata