sync-changelog

Pass

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes standard development commands including git, gh, npm, and pnpm to manage versioning and documentation builds.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its handling of data from the apps/kimi-code/CHANGELOG.md file.
  • Ingestion points: apps/kimi-code/CHANGELOG.md where changelog entries, commit hashes, and PR numbers are read.
  • Boundary markers: None present to distinguish data from command parameters.
  • Capability inventory: Shell access using git show and gh pr view with extracted identifiers.
  • Sanitization: No validation is performed on commit hashes or PR numbers before execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 7, 2026, 07:30 PM
Security Audit — agent-trust-hub — sync-changelog