The Agent Skills Directory

[SAFE]: The skill implements a workflow for creating local documentation files using restricted, non-networked tools (Read, Write, Glob, Grep). The tool usage is consistent with its stated purpose of managing .agents/skills/ directories.
[INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection because it extracts agent instructions from untrusted project files or user activity and persists them in the skill directory. * Ingestion points: Observations of repetitive patterns and workflows in current project files (SKILL.md Step 1). * Boundary markers: The template (Step 2) uses standard Markdown headers but does not implement specific delimiters or 'ignore' prefixes for user-provided data within the generated protocol. * Capability inventory: The skill utilizes the 'Write' tool to save new instruction sets to the '.agents/skills/' path. * Sanitization: No sanitization, escaping, or validation logic is defined to check the extracted content before it is finalized into a persistent skill file.

skillify