search-tools

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): The content consists of a decision tree and usage examples. No override patterns, jailbreak attempts, or instructions to ignore safety filters were detected.
  • [Data Exposure & Exfiltration] (SAFE): No sensitive file paths, hardcoded secrets, or unauthorized network operations are present. The mention of an API key for the 'Morph' tool is informational and does not include a credential.
  • [Remote Code Execution] (SAFE): The skill provides example CLI commands for local tools (leann, ast-grep). It does not perform remote downloads, pipe-to-shell operations, or use dynamic code execution functions.
  • [Indirect Prompt Injection] (LOW): This category is flagged because the skill instructs the agent to ingest and process data from external search tools.
  • Ingestion points: Search results from LEANN, AST-grep, Morph, and Grep (SKILL.md).
  • Boundary markers: None specified in this tool hierarchy guide.
  • Capability inventory: Shell execution of search commands (leann, grep) as documented in SKILL.md.
  • Sanitization: No sanitization logic is described in this documentation-focused skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:02 PM