semantic-code-search

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions specify the use of uvx cocoindex-code==0.1.11 to automatically fetch and launch the MCP server from the public Python Package Index (PyPI). This creates a runtime dependency on an external package from a vendor that is not in the trusted list.- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external codebase content.
  • Ingestion points: The mcp__cocoindex-code__search tool retrieves code snippets and documentation from the local environment via semantic search.
  • Boundary markers: There are no instructions defining delimiters or explicit warnings for the agent to ignore instructions embedded within the retrieved code.
  • Capability inventory: The agent uses the tool to explore unfamiliar codebases and locate implementations by natural language concept.
  • Sanitization: No sanitization or validation of the content retrieved from the codebase is performed before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 08:41 AM