The Agent Skills Directory

[PROMPT_INJECTION] (LOW): The skill exhibits a surface for Indirect Prompt Injection (Category 8) as it ingests untrusted data from a file without using delimiters or sanitization.\n- Ingestion points: Reads content from the file path provided as argument $1 in SKILL.md.\n- Boundary markers: Absent. The skill does not provide delimiters to separate external data from system instructions.\n- Capability inventory: Includes Read and Write tool permissions in SKILL.md.\n- Sanitization: Absent. No validation or filtering is applied to the input content before processing.

plan2json