skills/zzafergok/skills/skill-lookup/Gen Agent Trust Hub

skill-lookup

Warn

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches files, including scripts and documentation, from the external prompts.chat MCP server using the get_skill tool.
  • [REMOTE_CODE_EXECUTION]: The installation instructions direct the agent to save helper scripts (Python, shell, etc.) to the local .claude/skills/ directory, staging untrusted code for potential execution.
  • [COMMAND_EXECUTION]: The skill requires the agent to perform filesystem operations, such as creating directories and writing files, using content and paths derived from external tool outputs.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting and processing content from an external API without sanitization. Ingestion points: search_skills and get_skill tool outputs; Boundary markers: Absent; Capability inventory: File-write to sensitive local paths and potential execution of downloaded scripts; Sanitization: Absent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 31, 2026, 02:03 AM
Security Audit — agent-trust-hub — skill-lookup