which-skill

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill mentions an external website at mohitagw15856.github.io/pm-claude-skills/ for interacting with the tools. This domain belongs to the skill's author and uses a well-known service (GitHub Pages).
  • [COMMAND_EXECUTION]: Instructions direct the agent to utilize npx pm-claude-skills list for searching a catalog of available skills. This involves executing a package from the npm registry, which is an external dependency managed by the author.
  • [INDIRECT_PROMPT_INJECTION]: The skill identifies a surface for indirect injection as it processes user-provided task descriptions to perform routing. * Ingestion points: User input captured as The task in the user's own words. * Boundary markers: No delimiters or instructions are used to separate user data from the agent's logic. * Capability inventory: The agent is authorized to execute shell commands (npx) and read local files (SKILLS.md, WORKFLOWS.md). * Sanitization: No input sanitization or validation logic is defined to prevent embedded instructions in user tasks from influencing the routing outcome.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 05:56 PM
Security Audit — agent-trust-hub — which-skill