asset-aware-mcp-harness

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill defines routine development workflows for repository maintenance and code quality checks using local resources.
  • [COMMAND_EXECUTION]: The skill provides canonical commands for running linters, type checkers, and tests using uv, npm, and docker. These commands are standard for project development and are executed against the local repository context.
  • [DATA_EXFILTRATION]: No network exfiltration or sensitive file access is present. The skill manages local MCP configuration files (mcp.json, config.toml) in a way that respects existing user settings.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill instructs the agent to read and follow markdown-based workflows in .clinerules/workflows/ (e.g., /full-check.md).
  • Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are present in this harness.
  • Capability inventory: The skill utilizes subprocess execution for uv, npm, and docker for local validation tasks.
  • Sanitization: No sanitization is performed on the content of the workflow files before processing.
  • Analysis: This represents a standard surface for indirect prompt injection via local files, but is intended for the primary purpose of repository task automation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 08:24 PM
Security Audit — agent-trust-hub — asset-aware-mcp-harness