agents-md

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): The files contain templates and a refactoring prompt meant for AI agents. These instructions are designed to optimize developer workflows and do not attempt to bypass safety constraints, extract system prompts, or induce restricted behavior.
  • [Command Execution] (SAFE): The documentation and templates reference standard development toolchains (e.g., pnpm, npm, poetry, cargo, go). These commands are contextually appropriate for the project types described and do not use high-risk flags or attempt privilege escalation.
  • [External Downloads] (SAFE): While the templates mention package installation commands, they target official registries (npm, PyPI, Crates.io) via standard package managers. No suspicious third-party URLs or piped shell script executions (e.g., curl | bash) are present.
  • [Data Exposure & Exfiltration] (SAFE): The skill does not access sensitive local files (~/.ssh, etc.) or perform network operations. No hardcoded credentials or API keys were found in the templates.
  • [Obfuscation] (SAFE): All markdown files are written in plain text. No Base64 encoding, zero-width characters, or homoglyphs were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:05 PM