Skills
Navigate
skills/ravnhq/ai-toolkit/agent-skill-creator/Gen Agent Trust Hub

agent-skill-creator

Pass

Audited by Gen Agent Trust Hub on Oct 17, 2026

Risk Level: SAFEPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [PRIVILEGE_ESCALATION]: The scripts/init_skill.py script uses example_script.chmod(0o755) to set executable permissions on a newly created example script.
  • [DYNAMIC_EXECUTION]: The scripts/init_skill.py script generates several files including SKILL.md and scripts/example.py by populating hardcoded templates with user-provided variables.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user prompts to generate new skill instructions and structures.
  • Ingestion points: User input for skill name and description is interpolated into the generated SKILL.md.
  • Boundary markers: The skill relies on template structure but lacks explicit instructions to ignore potentially malicious content within user-supplied descriptions.
  • Capability inventory: The skill utilizes Write and Bash tools to create and manage the local skill directory structure.
  • Sanitization: The scripts/quick_validate.py utility validates naming conventions and description lengths to prevent malformed metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 17, 2026, 02:55 AM