skills/chipctx/skills/chip-specs/Gen Agent Trust Hub

chip-specs

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPERSISTENCEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill contains instructions for the AI agent to execute shell commands to configure its environment. Specifically, it directs the agent to run claude mcp add to add the author's MCP server to the platform configuration.
  • [EXTERNAL_DOWNLOADS]: The skill references and downloads configuration and components from vendor-controlled infrastructure, including the MCP server hosted at https://mcp.chipctx.com and the plugin repository at github.com/chipctx/skills.
  • [PERSISTENCE]: The recommended setup procedure involves modifying the platform's global configuration file (mcp.json), which ensures the vendor's MCP server remains active across agent sessions.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a flow that ingests data from various project files to automate chip identification.
  • Ingestion points: The skill automatically scans user message text, C/C++ header files (e.g., stm32f4xx.h), and build configuration files such as platformio.ini, Makefile, CMakeLists.txt, sdkconfig, and prj.conf.
  • Boundary markers: The instructions lack explicit delimiters or specific warnings to the agent to treat content from these external files as untrusted data.
  • Capability inventory: The agent can invoke several MCP tools to retrieve detailed technical data and possesses the capability to execute environment modification commands.
  • Sanitization: No sanitization or validation process is described for the strings extracted from project files before they are passed to the chip_identify tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 07:08 AM
Security Audit — agent-trust-hub — chip-specs