swarm-advanced
Advanced swarm orchestration patterns for research, development, testing, and complex distributed workflows
47
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillValidation for skill structure
Evaluation results
100%
46%Automated Research Orchestration Script
Research swarm topology and agent configuration
Mesh topology
0%
100%
Adaptive strategy
0%
100%
MaxAgents 6
0%
100%
Researcher agents
100%
100%
Analyst agents
100%
100%
Documenter agent
0%
100%
Parallel gather
40%
100%
Memory namespace
100%
100%
Memory TTL set
100%
100%
Knowledge graph storage
0%
100%
Report generation workflow
100%
100%
swarm_init called
100%
100%
Without context: $0.4817 · 2m 29s · 20 turns · 26 in / 9,387 out tokens
With context: $0.5670 · 2m 6s · 22 turns · 305 in / 6,962 out tokens
100%
77%Development Swarm for a New Microservice
Development swarm with fault tolerance and state management
Hierarchical topology
0%
100%
Balanced strategy
0%
100%
MaxAgents 8
0%
100%
Architect agent
100%
100%
Coder agents
42%
100%
Tester and reviewer agents
42%
100%
Sequential architecture phase
37%
100%
Parallel implementation
37%
100%
Fault tolerance setup
10%
100%
Error handling with recovery
20%
100%
State snapshot
11%
100%
Memory namespace for design
0%
100%
Without context: $0.7842 · 2m 55s · 22 turns · 25 in / 9,836 out tokens
With context: $0.5827 · 2m 16s · 20 turns · 25 in / 7,375 out tokens
100%
97%Comprehensive Test Automation Swarm
Testing swarm monitoring and performance analysis
Star topology
0%
100%
Parallel strategy
0%
100%
MaxAgents 7
0%
100%
Tester agents spawned
28%
100%
Parallel test execution
12%
100%
Monitor with interval
0%
100%
Quality assessment
0%
100%
Bottleneck analysis
0%
100%
Health check
0%
100%
Trend analysis
0%
100%
CLI fallback
0%
100%
Performance report
0%
100%
Without context: $0.6977 · 3m 37s · 22 turns · 73 in / 13,927 out tokens
With context: $0.7238 · 3m 43s · 24 turns · 186 in / 9,765 out tokens
- Evaluated
- Agent
- Claude Code
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.