product-discovery

Warn

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to scan the user's home directory for other skill installations (e.g., ~/.claude/skills/) using the ls command. Evidence: SKILL.md.
  • [EXTERNAL_DOWNLOADS]: The skill encourages the installation of remote packages using 'npx skills add' from an unverified external source. Evidence: README.md.
  • [PROMPT_INJECTION]: The skill processes untrusted research data but lacks sanitization or boundary markers to prevent indirect prompt injection. Ingestion: interview notes and support tickets in SKILL.md; Boundary: Absent; Capability: Filesystem listing and remote execution; Sanitization: Absent. Evidence: SKILL.md.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 7, 2026, 03:05 PM
Security Audit — agent-trust-hub — product-discovery