mailmap-cleanup

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes standard git commands (git log) to retrieve author names and emails from the repository's version history.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from git logs (contributor names and emails) which can be influenced by external actors who submit commits to the repository.\n
  • Ingestion points: The output of git log --format='%aN|%aE' | sort -u is analyzed by the agent (SKILL.md, Step 1 and Step 5).\n
  • Boundary markers: Absent; the raw output from the git command is processed directly by the agent without specific encapsulation.\n
  • Capability inventory: Executes command-line tools (git) and performs file writes to the .mailmap file.\n
  • Sanitization: Absent; the agent relies on its own logic and user confirmation to validate and format the ingested strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 06:25 AM
Security Audit — agent-trust-hub — mailmap-cleanup