skills/chann/skills/git-commit-push/Gen Agent Trust Hub

git-commit-push

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes git commands including git push and git push -u origin to facilitate synchronization of local changes with a remote repository.\n- [DATA_EXFILTRATION]: Mitigates data exfiltration risks through a comprehensive 'Red Flags' section that explicitly prohibits the commitment of sensitive files such as .env, private keys (.pem, .key), and credentials.\n- [PROMPT_INJECTION]: Presents a surface for indirect prompt injection by processing untrusted data from the local working tree during the commit workflow.\n
  • Ingestion points: Local file content and directory structures as referenced in the inspection workflow.\n
  • Boundary markers: Relies on the structured planning steps defined in the core git-commit skill to maintain context boundaries.\n
  • Capability inventory: Possesses filesystem and network capabilities via the git CLI for adding, committing, and pushing data.\n
  • Sanitization: Incorporates a blacklist of sensitive file patterns to prevent the unintentional exposure of secrets found within the working tree.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 01:12 PM