skill-creator

Pass

Audited by Gen Agent Trust Hub on May 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions in SKILL.md guide the agent to fetch content from arbitrary remote URLs using curl or web_fetch during the installation process.
  • [COMMAND_EXECUTION]: The skill includes Python scripts (scripts/init_skill.py, scripts/package_skill.py) that perform file system management, including creating directories, writing files, and modifying file permissions (chmod 755).
  • [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection by design, as it instructs the agent to ingest and follow instructions from external sources such as remote URLs and INSTALL.md files.
  • Ingestion points: Remote URLs and external skill directories in SKILL.md.
  • Boundary markers: Absent; the skill does not use specific delimiters or warnings to isolate instructions found in external skill packages.
  • Capability inventory: Includes bash, read, write, edit tools, and the execution of Python scripts for workspace management.
  • Sanitization: None; the skill does not validate or sanitize instructions contained within downloaded installation files.
Audit Metadata
Risk Level
SAFE
Analyzed
May 27, 2026, 02:23 AM
Security Audit — agent-trust-hub — skill-creator