checkpoint
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were detected in the instructions or configuration files.
- [DATA_EXPOSURE]: The skill is designed to inspect repository state (branch names, commit hashes, and file diffs) and project trackers. This access is necessary for its stated purpose of generating a work-in-progress summary.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the working environment (repository files and external trackers).
- Ingestion points: Inspects repository identity, branch, HEAD, and relevant files (SKILL.md).
- Boundary markers: None present to isolate ingested data from the prompt context.
- Capability inventory: Limited to reading file system state and writing a markdown file to a specific path (
~/.engineering-workflow/). No network or code execution capabilities are present. - Sanitization: None present. However, the risk is mitigated by the skill's lack of high-privilege capabilities and its explicit instruction to treat the checkpoint as historical and re-verify state before acting.
Audit Metadata