skills/entireio/cli/changelog/Gen Agent Trust Hub

changelog

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes git and gh CLI tools to retrieve repository information, including tags, merge commits, and pull request metadata. These operations are standard for change management tasks.\n- [PROMPT_INJECTION]: The skill ingests untrusted metadata from merged pull requests and commit history, which constitutes a surface for indirect prompt injection.\n
  • Ingestion points: Pull request titles via gh pr list, commit messages via git log, and team data from memory/project_team_members.md.\n
  • Boundary markers: Absent. No delimiters or instructions to ignore embedded commands are used when processing the data.\n
  • Capability inventory: Command execution (git, gh) and file system modification (CHANGELOG.md).\n
  • Sanitization: Absent. The skill does not sanitize or validate strings retrieved from external contributors before including them in the output.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 06:05 AM
Security Audit — agent-trust-hub — changelog