skill-developer

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • No Code (SAFE): The analyzed skill consists entirely of Markdown documentation and configuration schemas. No executable scripts, binaries, or automated installation commands are included in the file set.
  • Indirect Prompt Injection (SAFE): The documented architecture involves processing user prompts and local file contents, which represents an indirect prompt injection surface.
  • Ingestion points: The system ingests data via the UserPromptSubmit (user input) and PreToolUse (file content/tool inputs) hooks.
  • Boundary markers: The documentation recommends using ASCII-based formatting to distinguish skill suggestions and block messages from the rest of the agent's context.
  • Capability inventory: The system is capable of injecting context into the agent's prompt and blocking the execution of file-editing tools based on matched patterns.
  • Sanitization: The provided documentation does not detail specific sanitization or escaping mechanisms for the data being matched against regex patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:42 PM