skills/gildesmarais/dotfiles/dev/Gen Agent Trust Hub

dev

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses standard development tools such as rg (ripgrep) for code search and git for commit history and status. These operations are scoped to the project environment and used for context gathering during the 'Shared prep' and 'implement' phases.
  • [EXTERNAL_DOWNLOADS]: The skill integrates with Model Context Protocol (MCP) tools like user-dash-api and Context7 to retrieve documentation. These are platform-level integrations for context retrieval rather than arbitrary remote code downloads from untrusted sources.
  • [PROMPT_INJECTION]: Functional constraints are present (e.g., refusing to answer product strategy questions or inventing APIs), which serve to maintain the agent's focus on implementation and accuracy rather than bypassing safety or behavioral guardrails. The skill processes project-specific files like AGENTS.md and CONTEXT.md to guide its behavior, which is standard for development-oriented agents.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 09:30 AM
Security Audit — agent-trust-hub — dev