The Agent Skills Directory

[SAFE]: The skill's primary function involves scanning local project-specific files such as agents/*.md, skills/*/SKILL.md, rules/**/*.md, and .mcp.json to estimate token counts and line lengths. This is consistent with its stated purpose of auditing context budget.
[SAFE]: No network operations, data exfiltration patterns, or attempts to access sensitive system directories (e.g., .ssh, .aws) were detected.
[SAFE]: The skill does not include any remote code execution, dynamic code generation, or obfuscation techniques. All operations are local and analytical in nature.
[SAFE]: The logic provided for identifying 'redundant' or 'bloated' components is based on static analysis of file contents and lengths, presenting no risk to the integrity of the environment.

context-budget