The Agent Skills Directory

[SAFE]: The skill enforces a strict read-only policy, prohibiting mutations such as running scans, editing files, or modifying configurations. It includes explicit instructions to avoid collecting or echoing secrets, tokens, or full configuration files. It also implements an evidence-based diagnostic workflow that treats command output and repository files as data rather than instructions.
[COMMAND_EXECUTION]: Employs the endorctl api CLI and jq for troubleshooting. These tools are restricted to read-only diagnostic lookups (get/list) and are used to gather evidence for identifying root causes.
[EXTERNAL_DOWNLOADS]: Provides links to official Endor Labs documentation for user reference and repair guidance.

endor-troubleshooter