changelog-writer

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from git commit messages, pull request titles, and issue descriptions.
  • Ingestion points: Data is ingested from git history logs, PR metadata, and issue references in SKILL.md.
  • Boundary markers: No specific delimiters or "ignore embedded instructions" markers are defined to isolate input data from the prompt logic.
  • Capability inventory: The skill utilizes shell command execution (git, npm, gh, bash) to perform its core tasks.
  • Sanitization: No explicit sanitization or filtering of commit messages or PR content is described in the workflow.
  • [COMMAND_EXECUTION]: The skill requires the execution of various shell commands and custom scripts to retrieve version history and automate the release process.
  • Evidence: Usage of git log, git describe, and gh release create commands, along with instructions for installing conventional-changelog-cli and git-cliff.
  • Context: These operations are standard practices for software release automation and match the skill's primary function.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:13 PM
Security Audit — agent-trust-hub — changelog-writer