The Agent Skills Directory

[SAFE]: The skill is designed to perform a static audit of local project files (such as agents, skills, rules, and MCP configurations) to estimate token consumption.
[SAFE]: No network operations, external downloads, or remote code execution patterns were identified.
[SAFE]: The skill does not attempt to override agent behavior or bypass safety guidelines; its instructions are focused entirely on the auditing and reporting of metadata.
[SAFE]: No obfuscation or credential harvesting techniques were detected. The access to .mcp.json and project files is limited to counting components and estimating schema sizes for performance reporting.

context-budget