create-agent-skills

Pass

Audited by Gen Agent Trust Hub on Mar 2, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill documents and provides workflows for executing shell commands and scripts (Bash, Python, Node.js) to automate tasks. These operations are restricted to the local environment and follow explicit user-guided workflows.
  • [EXTERNAL_DOWNLOADS]: Instructions include installing standard, well-known libraries (e.g., pdfplumber, pypdf, gh CLI) from official registries. References point to trusted documentation sites and the official anthropics/skills repository.
  • [CREDENTIALS_UNSAFE]: The skill contains a dedicated reference file that explicitly teaches how to avoid exposing API keys in chat by using a local wrapper script and environment variables stored in .env files.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines patterns for data ingestion and mitigates risks by mandating boundary markers (XML tags), success criteria, and validation loops for all generated content.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 2, 2026, 04:13 AM