The Agent Skills Directory

[SAFE]: The skill's primary function is to modify local instruction files (AGENTS.md or CLAUDE.md) to refine agent behavior. This is a standard project management task and is performed without accessing sensitive system paths or credentials.
[SAFE]: No network operations, external downloads, or remote code execution patterns were detected. The skill operates entirely within the local file system scope.
[SAFE]: There is no evidence of obfuscation, multi-layer encoding, or hidden instructions. The logic is transparent and focused on structured data categorization (Always/Ask/Never).
[SAFE]: While the skill ingests user input and writes it to instruction files (Category 8 surface), this is the explicit and intended purpose of the tool. The 'additive-only' constraint prevents the accidental or malicious deletion of existing security boundaries.

refine-boundaries