leann-search

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to interact with the local system via the leann CLI tool using the Bash tool. It provides examples for searching (leann search), listing indices (leann list), and rebuilding the index (leann build).
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes content from the local codebase.
  • Ingestion points: The tool reads and indexes content from project directories (src, tests, scripts) and returns excerpts during semantic searches.
  • Boundary markers: No specific boundary markers or instructions are provided to help the agent distinguish between search tool metadata and potentially malicious instructions embedded in the indexed files.
  • Capability inventory: The skill utilizes the Bash and Read tools to perform operations.
  • Sanitization: There is no evidence of sanitization or filtering of the content retrieved from the codebase files before it is presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 07:19 AM
Security Audit — agent-trust-hub — leann-search