The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content from external sources which could contain malicious instructions. 1. Ingestion points: Data retrieved via WebFetch and WebSearch tools as defined in SKILL.md. 2. Boundary markers: The instructions lack explicit delimiters to isolate external data from the prompt. 3. Capability inventory: The skill has access to Read, Glob, Grep, and restricted Bash tools in SKILL.md. 4. Sanitization: No sanitization is performed on external inputs.

ask