Skills
skills/outlinedriven/odin-codex-plugin/mcp-builder/Gen Agent Trust Hub

mcp-builder

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION] (SAFE): The file scripts/connections.py implements an MCP client capable of executing local commands via the stdio transport, which is standard for local MCP tools.
  • [EXTERNAL_DOWNLOADS] (SAFE): The SKILL.md instructions direct the agent to fetch documentation and SDK READMEs from reputable external sources (github.com and modelcontextprotocol.io) to ensure accurate implementation.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill has a surface for indirect injection due to its ingestion of external data coupled with command execution capabilities. 1. Ingestion points: Documentation URLs in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Command execution via connections.py and build scripts. 4. Sanitization: None.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:41 PM