How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

system-health-audit - Prompt

Name: system-health-audit
Author: cdalsoniii

Prompt file imported from cdalsoniii/brightpath-coder (.cursor/commands/system-health-audit.md). Copyright stays with the author.

System Health Audit

Verify that the entire agent/skill/rule/command ecosystem is healthy and all 20 concerns have active coverage.

Steps

Agent Health Check:
- List all agents in .cursor/agents/
- For each agent, verify:
  - Trigger patterns match existing file paths in the codebase
  - Referenced skills exist in .cursor/skills/
  - Referenced rules exist in .cursor/rules/
  - Referenced plans exist in .cursor/plans/
- Report orphaned references
Skill Health Check:
- List all skills in .cursor/skills/
- For each skill, verify:
  - SKILL.md exists and has valid frontmatter
  - Referenced rules exist
  - Glob patterns are not empty
- Report unused skills (not referenced by any agent)
Rule Health Check:
- List all rules in .cursor/rules/
- For each rule, verify:
  - Valid frontmatter (description, globs)
  - Glob patterns match existing files (if file-scoped)
- Report orphaned rules (not referenced by any agent or skill)
Command Health Check:
- List all commands in .cursor/commands/
- For each command, verify:
  - Valid frontmatter
  - Referenced rules/skills/agents exist
- Report orphaned commands
Concern Coverage Matrix:
- Using rule 130 (concern-trigger-registry), verify each of the 20 concerns:
  - Has at least one agent with matching trigger patterns
  - Has at least one verification command
  - Has at least one backing skill or rule
- Report uncovered concerns
Overlap Detection:
- Identify agents with identical trigger patterns (potential duplication)
- Verify differentiated agents (code-review-specialist vs code-reviewer, etc.) have distinct trigger scopes

Generate Health Report:

# System Health Audit — [date]

## Summary
- Agents: N total, N healthy, N with issues
- Skills: N total, N referenced, N orphaned
- Rules: N total, N applied, N orphaned
- Commands: N total
- Concerns: N/20 fully covered

## Issues
...

## Recommendations
...

References

.cursor/rules/130-concern-trigger-registry.mdc

Install system-health-audit and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

# System Health Audit Verify that the entire agent/skill/rule/command ecosystem is healthy and all 20 concerns have active coverage. ## Steps 1. **Agent Health Check**: - List all agents in `.cursor/agents/` - For each agent, verify: - Trigger patterns match existing file paths in the codebase - Referenced skills exist in `.cursor/skills/` - Referenced rules exist in `.cursor/rules/` - Referenced plans exist in `.cursor/plans/` - Report orphaned references 2. **Skill Health Check**: - List all skills in `.cursor/skills/` - For each skill, verify: - SKILL.md exists and has valid frontmatter - Referenced rules exist - Glob patterns are not empty - Report unused skills (not referenced by any agent) 3. **Rule Health Check**: - List all rules in `.cursor/rules/` - For each rule, verify: - Valid frontmatter (description, globs) - Glob patterns match existing files (if file-scoped) - Report orphaned rules (not referenced by any agent or skill) 4. **Command Health Check**: - List all commands in `.cursor/commands/` - For each command, verify: - Valid frontmatter - Referenced rules/skills/agents exist - Report orphaned commands 5. **Concern Coverage Matrix**: - Using rule 130 (concern-trigger-registry), verify each of the 20 concerns: - Has at least one agent with matching trigger patterns - Has at least one verification command - Has at least one backing skill or rule - Report uncovered concerns 6. **Overlap Detection**: - Identify agents with identical trigger patterns (potential duplication) - Verify differentiated agents (code-review-specialist vs code-reviewer, etc.) have distinct trigger scopes 7. **Generate Health Report**: ```markdown # System Health Audit — [date] ## Summary - Agents: N total, N healthy, N with issues - Skills: N total, N referenced, N orphaned - Rules: N total, N applied, N orphaned - Commands: N total - Concerns: N/20 fully covered ## Issues ... ## Recommendations ... ``` ## References - `.cursor/rules/130-concern-trigger-registry.mdc`

{ "prompt": "# System Health Audit\n\nVerify that the entire agent/skill/rule/command ecosystem is healthy and all 20 concerns have active coverage.\n\n## Steps\n\n1. **Agent Health Check**:\n - List all agents in `.cursor/agents/`\n - For each agent, verify:\n - Trigger patterns match existing file paths in the codebase\n - Referenced skills exist in `.cursor/skills/`\n - Referenced rules exist in `.cursor/rules/`\n - Referenced plans exist in `.cursor/plans/`\n - Report orphaned references\n\n2. **Skill Health Check**:\n - List all skills in `.cursor/skills/`\n - For each skill, verify:\n - SKILL.md exists and has valid frontmatter\n - Referenced rules exist\n - Glob patterns are not empty\n - Report unused skills (not referenced by any agent)\n\n3. **Rule Health Check**:\n - List all rules in `.cursor/rules/`\n - For each rule, verify:\n - Valid frontmatter (description, globs)\n - Glob patterns match existing files (if file-scoped)\n - Report orphaned rules (not referenced by any agent or skill)\n\n4. **Command Health Check**:\n - List all commands in `.cursor/commands/`\n - For each command, verify:\n - Valid frontmatter\n - Referenced rules/skills/agents exist\n - Report orphaned commands\n\n5. **Concern Coverage Matrix**:\n - Using rule 130 (concern-trigger-registry), verify each of the 20 concerns:\n - Has at least one agent with matching trigger patterns\n - Has at least one verification command\n - Has at least one backing skill or rule\n - Report uncovered concerns\n\n6. **Overlap Detection**:\n - Identify agents with identical trigger patterns (potential duplication)\n - Verify differentiated agents (code-review-specialist vs code-reviewer, etc.) have distinct trigger scopes\n\n7. **Generate Health Report**:\n ```markdown\n # System Health Audit — [date]\n \n ## Summary\n - Agents: N total, N healthy, N with issues\n - Skills: N total, N referenced, N orphaned\n - Rules: N total, N applied, N orphaned\n - Commands: N total\n - Concerns: N/20 fully covered\n \n ## Issues\n ...\n \n ## Recommendations\n ...\n ```\n\n## References\n- `.cursor/rules/130-concern-trigger-registry.mdc`", "variables": [], "notes": "Comprehensive system health audit — verify all 20 concerns have active agent coverage, working triggers, valid skills, and passing verification commands Mode: agent. Tools: read, search, glob, shell.", "metadata": { "source": { "provider": "github-cursor-commands", "repository": "https://github.com/cdalsoniii/brightpath-coder", "path": ".cursor/commands/system-health-audit.md", "ref": "6fab58a589619342c36cd1f983a53f36671c7a20", "url": "https://github.com/cdalsoniii/brightpath-coder/blob/6fab58a589619342c36cd1f983a53f36671c7a20/.cursor/commands/system-health-audit.md", "key": "cdalsoniii/brightpath-coder/.cursor/commands/system-health-audit.md" } }, "model_hints": [ "claude-sonnet" ] }

system-health-audit

About

System Health Audit

Steps

References

Install system-health-audit and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Use it

Manifest

Reviews