write-a-skill

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a development tool for authoring and structuring agent skills. It includes instructional content, workflow patterns (Sequential, Iterative, etc.), and technical guidelines.
  • [SAFE]: No malicious patterns such as prompt injection, obfuscation, or data exfiltration were detected. Instructions focus on standard software development and documentation practices.
  • [SAFE]: Mentions of external services (GitHub, NPM, PyPI) and organizations (Anthropic) are used in a benign, informative context. Examples of scripts (e.g., Python validation scripts) use standard libraries and do not perform dangerous operations.
  • [SAFE]: While the skill mentions reading a local CONTEXT.md file to align with project terminology, this is a standard practice for project-aware tools and does not present an active security risk in the context of this skill's logic.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 06:24 PM
Security Audit — agent-trust-hub — write-a-skill