Skills
skills/ruvnet/ruflo/harness-mcp-scan/Gen Agent Trust Hub

harness-mcp-scan

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the metaharness utility from the official NPM registry via the npx command.
  • [COMMAND_EXECUTION]: Uses the Bash tool to run shell commands for package installation and static analysis.
  • [REMOTE_CODE_EXECUTION]: Downloads and executes code from a remote source at runtime using npx -p metaharness@latest.
  • [PROMPT_INJECTION]: The skill processes untrusted data from project configuration files, which presents a surface for indirect prompt injection.
  • Ingestion points: Reads .mcp/servers.json and .harness/claims.json via an external script scripts/mcp-scan.mjs.
  • Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present in the documentation.
  • Capability inventory: The skill has access to shell execution via the Bash tool.
  • Sanitization: No validation or sanitization of the input file content is documented before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 06:39 PM
Security Audit — agent-trust-hub — harness-mcp-scan