release-notes

Pass

Audited by Gen Agent Trust Hub on Oct 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on a Node.js script to execute local git and gh (GitHub CLI) commands. These are used to determine the repository root, parse version blocks from CHANGELOG.md, fetch commit history, and query the GitHub GraphQL API. The implementation uses execFile and execFileSync, which avoid shell evaluation and mitigate common command injection vulnerabilities.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes data from potentially untrusted sources to generate the release body, creating a surface for indirect prompt injection.\n
  • Ingestion points: The skill reads and parses CHANGELOG.md, Git commit history, and PR metadata retrieved from the GitHub API.\n
  • Boundary markers: No specific delimiters or instructions to ignore embedded malicious content are present in the script logic.\n
  • Capability inventory: The agent is authorized to execute repository-level commands and write to the local file system (v<version>.md).\n
  • Sanitization: The script includes regex validation for PR numbers (\\d+) and uses non-shell process execution, providing robust protection against most forms of injection into the automation tool itself.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 28, 2026, 10:43 PM