analyze-plugin

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Node.js validation script (node scripts/plugin-routing-validate.mjs) to ensure the generated artifacts meet the required schema and logic gates.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted content from third-party plugins (READMEs, manifests, and instructions) to decide how to reimplement them.
  • Ingestion points: Reads various files from ~/.claude/plugins/cache/, including Markdown documentation, JSON manifests, and skill definitions.
  • Boundary markers: The instructions lack explicit boundary markers or directives to treat analyzed content as data rather than instructions.
  • Capability inventory: The skill has access to Bash, Read, Write, Edit, and Skill tools, enabling it to write persistent artifacts to the repository and execute scripts.
  • Sanitization: There is no evidence of sanitization, filtering, or validation performed on the text content of the analyzed plugins before it influences the agent's decision-making logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 06:27 AM
Security Audit — agent-trust-hub — analyze-plugin