skill-writer

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill guides the agent to use shell commands such as mkdir to set up the directory structure for new skills. This behavior is expected for a developer tool and is confined to the specific scope of skill creation.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user requirements and project details to generate documentation and instructions for other skills. This creates a potential attack surface where malicious user input could be propagated into the generated skill's prompts.
  • Ingestion points: User-provided descriptions and specifications for new skills.
  • Boundary markers: The templates provided in the instructions use standard Markdown formatting but do not include specific delimiters or warnings to ignore potentially malicious instructions in the user-supplied content.
  • Capability inventory: The skill is designed to perform file system operations, including directory creation and file writing.
  • Sanitization: There is no mention of sanitizing or validating user input before it is used to generate the new skill's content.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 11:03 PM
Security Audit — agent-trust-hub — skill-writer