Skills
skills/remix-run/remix/make-change-file/Gen Agent Trust Hub

make-change-file

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands pnpm changes:preview and pnpm run lint for verifying changelog output and maintaining code standards.
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests untrusted data from PR diffs and commit messages while possessing write and execution capabilities.
  • Ingestion points: The agent reads PR diffs, commit ranges, and repository files like package.json (SKILL.md).
  • Boundary markers: No instructions for delimiters or ignoring embedded instructions in the ingested data are specified.
  • Capability inventory: The agent can write markdown change files to packages/*/.changes/ and execute pnpm shell commands.
  • Sanitization: No sanitization or validation of external content from PRs or commits is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 12:52 AM