ship-with-docs
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data from git logs and diffs, creating an attack surface for indirect prompt injection where malicious instructions in code or commit history could influence documentation output.
- Ingestion points: The skill uses
git diffandgit logto map changes as described inSKILL.md. - Boundary markers: No explicit delimiters are specified to isolate untrusted branch content from the agent's logic.
- Capability inventory: The skill possesses file-write capabilities for
CONTEXT.mdand ADR files, as well asgit commitauthority. - Sanitization: No input sanitization is performed. This is mitigated by the mandatory human-in-the-loop 'Batch Review' and 'Commit Gate' processes which require explicit confirmation before any action is taken.
Audit Metadata