The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data from the local project environment, creating a surface for indirect prompt injection. \n- Ingestion points: Reads project files via Read, Grep, and Glob, and git history via Bash(git:log). \n- Boundary markers: No delimiters or instructions to ignore embedded commands are present in the skill definition. \n- Capability inventory: The skill uses file access and shell tools to execute npm and git commands. \n- Sanitization: There is no defined process to sanitize or validate the content gathered from the project files.

metrics