skills/xltxai/skills/mcp-law-search/Gen Agent Trust Hub

mcp-law-search

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill configuration and instructions direct the agent to interact with an external MCP server at https://mcp.aixllaw.com/mcp. This server is the primary source for legal search results and is consistent with the skill vendor's infrastructure.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes legal data retrieved from an external API, which represents a potential attack surface if the retrieved data were to contain malicious instructions. However, the risk is mitigated by specific instructions that format and structure the output.\n
  • Ingestion points: The skill ingests text and record data from the search_law tool output.\n
  • Boundary markers: The skill defines a standard markdown response template to provide structure, though it does not use explicit security delimiters around the tool's raw output.\n
  • Capability inventory: The agent utilizes the search_law network-based tool and provides users with configuration snippets for local setup.\n
  • Sanitization: No specific content sanitization or instruction-filtering is applied to the legal records before presentation.\n- [COMMAND_EXECUTION]: The skill includes instructions for the agent to provide the user with specific JSON configuration blocks to be manually added to local configuration files (e.g., ~/.codebuddy/mcp.json). This facilitates the setup of the required MCP service through user-guided actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 08:05 AM
Security Audit — agent-trust-hub — mcp-law-search