skills/xmcp-dev/skills/create-tool/Gen Agent Trust Hub

create-tool

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • Dynamic Execution (LOW): The skill generates executable TypeScript (.ts) and React (.tsx) files from predefined templates. Although this is the primary purpose of the skill, generating scripts at runtime is a known security surface.
  • Indirect Prompt Injection (LOW):
  • Ingestion points: The skill collects untrusted data from the user via AskUserQuestion, including the tool-name, parameter names, and descriptions.
  • Boundary markers: Absent. The instructions do not specify any delimiters or safety warnings for the agent to use when interpolating user input into the code templates.
  • Capability inventory: The agent has the capability to write files to the local file system (specifically src/tools/).
  • Sanitization: Absent. There is no instruction to validate or sanitize the tool-name to prevent path traversal (e.g., providing ../../filename) or to escape strings within the generated code, which could lead to code injection in the resulting tool.
  • Command Execution (SAFE): The skill suggests the user run pnpm build. This is a standard and expected operation for the intended development workflow and does not pose an inherent risk in this context.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 11:14 AM