### Install Agent Orchestrator from URL

Source: https://github.com/mrclrchtr/skills/blob/main/README.md

Installs the 'agent-orchestrator' skill directly from its directory path on GitHub. This method allows for installation without cloning the repository locally.

```bash
npx skills add https://github.com/mrclrchtr/skills/tree/main/skills/agent-orchestrator
```

--------------------------------

### Install Agent Orchestrator Skill using npm

Source: https://context7.com/mrclrchtr/skills/llms.txt

Installs the agent-orchestrator skill from the specified repository or path. Supports installation from a named repository, a direct GitHub URL, or a local checkout. The `--list` flag can be used to see available skills without installing.

```bash
npx skills add mrclrchtr/skills --skill agent-orchestrator
npx skills add mrclrchtr/skills --list
npx skills add https://github.com/mrclrchtr/skills/tree/main/skills/agent-orchestrator
npx skills add . --skill agent-orchestrator
```

--------------------------------

### Install Agent Orchestrator from Local Checkout

Source: https://github.com/mrclrchtr/skills/blob/main/README.md

Installs the 'agent-orchestrator' skill from a local checkout of the repository. This is useful when working with a local copy of the skills project.

```bash
npx skills add . --skill agent-orchestrator
```

--------------------------------

### Install Agent Orchestrator Skill

Source: https://github.com/mrclrchtr/skills/blob/main/README.md

Installs the 'agent-orchestrator' skill using the 'skills' CLI from 'vercel-labs/skills'. This command adds the specified skill to your project's available skills.

```bash
npx skills add mrclrchtr/skills --skill agent-orchestrator
```

--------------------------------

### List Available Skills in Repository

Source: https://github.com/mrclrchtr/skills/blob/main/README.md

Lists all skills available within the 'mrclrchtr/skills' repository without installing them. This is useful for discovering available functionalities before installation.

```bash
npx skills add mrclrchtr/skills --list
```

--------------------------------

### Manage Workflow Plan Phases in YAML

Source: https://context7.com/mrclrchtr/skills/llms.txt

Tracks the progress of a multi-phase workflow. This YAML structure lists distinct phases of a project, their status (completed, in_progress, pending), and recommends maintaining exactly one step in progress at a time.

```yaml
phases:
  - name: "Audit authentication subsystem"
    status: completed
  - name: "Reconcile audit findings"
    status: completed
  - name: "Implement token generation cluster"
    status: in_progress
  - name: "Review and validate changes"
    status: pending
  - name: "Integration testing"
    status: pending
  - name: "Deliver completion report"
    status: pending
```

--------------------------------

### Decompose Work into Subsystems using YAML

Source: https://context7.com/mrclrchtr/skills/llms.txt

Splits a complex task into independently auditable subsystems, defining invariants for each. This YAML structure helps in managing complexity by outlining the components and their expected behaviors or properties.

```yaml
subsystems:
  - name: "API Surface"
    invariants:
      - All endpoints return consistent error formats
      - Authentication headers validated on protected routes
      - Rate limiting applied uniformly
  - name: "Core Logic"
    invariants:
      - Token generation uses cryptographically secure random
      - Expiration times configurable via environment
      - Refresh tokens single-use only
  - name: "Error Handling"
    invariants:
      - No sensitive data in error messages
      - All errors logged with correlation IDs
  - name: "Tests"
    invariants:
      - Coverage above 80% for new code
      - Integration tests for all auth flows
```

--------------------------------

### Define Workflow Scope and Success Criteria in YAML

Source: https://context7.com/mrclrchtr/skills/llms.txt

Defines the mission objectives, constraints, and success criteria for an agent's task. This YAML structure outlines the overall goal, limitations, expected outcomes, and authoritative sources for the task.

```yaml
mission: "Migrate user authentication from session-based to JWT tokens"
constraints:
  - Maintain backward compatibility for 30 days
  - No changes to public API signatures
  - All existing tests must pass
success_criteria:
  - JWT tokens issued on login
  - Token refresh endpoint operational
  - Session fallback available via feature flag
authoritative_sources:
  - docs/auth-spec.md
  - RFC 7519 (JWT)
```

--------------------------------

### Agent Template: Auditor Task Description

Source: https://context7.com/mrclrchtr/skills/llms.txt

Defines the task for an auditor agent, focusing on identifying issues within a specific subsystem without proposing solutions. It outlines the expected output format, including issue title, severity, evidence, and expected vs. actual results.

```markdown
# Auditor Task for API Surface Subsystem

Task:
- Audit the `API Surface` subsystem independently.
- Do not propose fixes yet; identify issues only.
- If a specialized skill is relevant to the subsystem, invoke it and follow its audit/checklist guidance.

Output (bullet list):
- issue title
- severity: critical/high/medium/low
- evidence: repo file + symbol (and line if stable)
- deterministic repro (commands/steps) or reasoning for why repro is not needed
- expected vs actual
- violated invariant (if known) or propose a new invariant

# Example Auditor Output
- **Missing rate limit on /api/refresh endpoint**
  - severity: high
  - evidence: src/routes/auth.ts:validateRefresh (line 142)
  - repro: `for i in {1..1000}; do curl -X POST /api/refresh; done`
  - expected: 429 after 100 requests; actual: all 1000 succeed
  - invariant violated: "Rate limiting applied uniformly"
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.