The Agent Skills Directory

[SAFE]: The skill operates as a transparent diagnostic utility. Its access to sensitive configuration files and environment data is restricted to identifying system health risks and security misconfigurations. It features a built-in privacy guard that redacts credential-like patterns (API keys, tokens, secrets) from collected data before reporting findings to the agent or user.
[COMMAND_EXECUTION]: The diagnostic process involves the execution of local shell and Node.js scripts to gather environmental metrics, service statuses, and resource usage. These commands are executed locally and are central to the skill's primary function as a system doctor.
[EXTERNAL_DOWNLOADS]: Documentation within the skill provides manual fix guidance for installing prerequisites like Node.js and the Clawhub CLI from well-known and vendor-trusted sources. These are recommendations for user-confirmed actions and do not involve automated background downloads.
[PROMPT_INJECTION]: The skill processes untrusted system data, presenting a surface for indirect prompt injection.
Ingestion points: Reads system logs (gateway.log), configuration JSON (openclaw.json), and workspace identity documents (agent.md, user.md).
Boundary markers: Present; the analysis process is governed by strict, domain-specific diagnostic protocols and thresholds.
Capability inventory: Extensive local command execution and file system access for system state inspection and report archival.
Sanitization: Employs regex-based redaction to filter and mask sensitive credential strings during the security audit phase.

botlearn-healthcheck