Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Coordinates multiple specialized agents to profile, optimize, and monitor multi-agent system performance. Covers context window compression, parallel execution, LLM cost tracking, and latency reduction across database, application, and frontend layers.
Combines profiling, cost tracking, and workload distribution into a single coordinated framework rather than requiring separate tools for each concern.
When to use it
- Profiling a slow API pipeline to find coordination bottlenecks
- Reducing monthly LLM token costs across a production agent fleet
- Running database, app, and frontend profiling agents in parallel
- Setting up automated performance feedback loops for agent workflows
- Validating orchestration changes with regression tests before deployment
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: optimize my e-commerce API pipeline with a $500/month LLM budget
- 1Profile database, application, and frontend agents to collect baseline metrics
- 2Identify coordination bottlenecks and high-cost LLM calls in the workflow
- 3Apply context compression and select cost-optimal models per task complexity
- 4Redistribute workloads using parallel execution and priority queuing
- 5Validate changes with regression tests and log before/after performance data
Performance report showing latency reduction, token usage savings, and throughput improvements with rollback checkpoints