skills/getsentry/skills/commit/Gen Agent Trust Hub

commit

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses standard Git commands (git branch, git commit) to check the environment and perform version control actions. These are benign and necessary for the stated purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user code changes and issue references to generate commit messages. It explicitly mitigates security risks by instructing the agent to never include secrets, PII, or support ticket contents in the final output. The ingestion points are the local repository and issue trackers, with capability limited to local Git commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 04:05 PM
Security Audit — agent-trust-hub — commit