git-master
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill inherently processes untrusted data from Git logs, diffs, and blame outputs, which constitutes a potential injection surface.\n
- Ingestion points:
git status,git diff,git log,git show, andgit blamecommands specified inSKILL.md.\n - Boundary markers: The skill instructs the agent to be conservative and evidence-led, but does not provide explicit delimiters for git data.\n
- Capability inventory: Shell command execution for git management (
git commit,git rebase,git branch) as defined inSKILL.md.\n - Sanitization: No explicit sanitization of git output is specified, but the skill mandates rigorous verification and safety checks before any write operations.\n- [SAFE]: The skill implements protective measures against accidental data exposure by instructing the agent not to log credentials, cookies, or internal upload headers when using GitHub for image attachments.
Audit Metadata