context-health
Use when validating or diagnosing the claude-ctx environment (consistency, duplicates, redundancy, optimization) - provides a diagnostic workflow and reporting structure for context health checks.
When & Why to Use This Skill
The Context Health skill provides a specialized diagnostic framework designed to maintain the integrity and efficiency of Claude's context environment. It streamlines the process of identifying configuration inconsistencies, detecting redundant data, and optimizing the overall health of the agent's workspace to ensure reliable and high-performance AI interactions.
Use Cases
- Post-Update Validation: Use this skill to verify system health and configuration integrity immediately after making changes to agent rules, modes, or environment settings.
- Performance Troubleshooting: Diagnose and resolve unexpected agent behavior by surfacing context duplicates or conflicting instructions that may degrade response quality.
- Context Optimization & Maintenance: Perform periodic health checks to identify and remove redundant information, helping to reduce token consumption and improve processing speed.
| name | context-health |
|---|---|
| description | Use when validating or diagnosing the cortex environment (consistency, duplicates, redundancy, optimization) - provides a diagnostic workflow and reporting structure for context health checks. |
Context Health
Overview
Provide a consistent diagnostic workflow for cortex environments: verify config integrity, detect duplicates, surface redundancies, and report actionable issues.
When to Use
- Checking system health after changes to agents, modes, or rules
- Troubleshooting unexpected cortex behavior
- Performing periodic maintenance checks
Avoid when:
- The task is unrelated to cortex configuration or context
Quick Reference
| Task | Load reference |
|---|---|
| Environment diagnostics | skills/context-health/references/doctor.md |
Workflow
- Confirm target scope and environment.
- Load the diagnostics reference.
- Run checks and gather findings.
- Summarize issues by severity.
- Recommend fixes and next steps.
Output
- Diagnostic report with severity levels
- Recommended fixes or follow-up actions
Common Mistakes
- Running fixes without reviewing findings
- Skipping context scope verification