fastify-plugins

Pass

Audited by Gen Agent Trust Hub on Oct 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external and potentially untrusted repository content, such as code, tests, and pull requests, which could contain malicious instructions. However, the risk is inherent to the skill's purpose as a development and review assistant.
  • Ingestion points: Reads repository files, contribution rules, linter configurations, and pull request diffs (SKILL.md, references/code.md).
  • Boundary markers: Explicitly instructs the agent to read local contribution rules and inspect project-specific configuration to establish authoritative context (SKILL.md).
  • Capability inventory: The skill performs file system reads and executes standard development lifecycle commands via npm run (lint, test, build) as documented in SKILL.md and references/test.md.
  • Sanitization: Relies on standard agent environment constraints; no additional automated sanitization of ingested code is specified before analysis.
  • [SAFE]: No security issues detected. The skill promotes secure development practices such as using the node: protocol for core modules and avoiding hardcoded secrets in favor of OIDC trusted publishing (references/code.md, references/actions.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 17, 2026, 07:36 AM