skill-creator

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs legitimate file system operations and executes local scripts to facilitate the creation and validation of new skills. It uses the Bash tool to run commands like ls for searching local directories and mkdir for scaffolding project structures. It also invokes its own utility scripts, scripts/validate-runtime.sh and tests/smoke.sh, to ensure generated content meets format requirements.
  • [EXTERNAL_DOWNLOADS]: The skill is configured to use the WebFetch tool, which is intended for searching external repositories (such as GitHub) to identify existing skills and avoid duplication. This aligns with the skill's primary purpose and involves user-directed discovery rather than automated remote code execution.
  • [SAFE]: The provided bash scripts follow security best practices, such as using set -euo pipefail and #!/usr/bin/env bash. The validation logic in scripts/validate-runtime.sh uses standard text processing tools (awk, sed) to parse YAML frontmatter safely without using dynamic evaluation or insecure execution paths.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 08:37 AM
Security Audit — agent-trust-hub — skill-creator