The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing content from external URLs and incorporating findings into instructions intended for other AI agents.
Ingestion points: The skill takes a user-provided domain and fetches data from the isitagentready.com API, which in turn scans the target website's headers, robots.txt, and other metadata.
Boundary markers: No specific boundary markers or sanitization logic is defined to separate external 'issue' messages from the rest of the generated prompt templates.
Capability inventory: The skill uses curl to perform network operations and provides templates for generating implementation reports and 'Fix Prompts' which are designed to be copy-pasted into coding agents.
Sanitization: The skill lacks sanitization of the {issue} and {message} fields retrieved from the external scan API before they are interpolated into the 'How to Implement' prompt blocks in SKILL.md.
[COMMAND_EXECUTION]: The skill provides instructions and examples for executing shell commands (curl) to interact with the scanner API and implement various agent protocols. While these are intended for manual execution or developer automation, they represent a capability that requires user oversight when handling URLs.
Evidence: Multiple curl examples are provided in README.md and SKILL.md for both scanning sites and validating implementations via the API endpoint.

agent-ready-cloudflare