swarm-orchestration

Name: swarm-orchestration
Author: spencermarx

301stars24forks

How to install swarm-orchestration?

spencermarx/open-code-review/swarm-orchestration

$npx -y skills add spencermarx/open-code-review --skill swarm-orchestration

Installs into the current project.

Summary

Orchestrate multi-agent swarms with agentic-flow for parallel task execution, dynamic topology, and intelligent coordination. Use when scaling beyond single agents, implementing complex workflows, or building distributed AI systems.

Files · 1

View on GitHub

SKILL.md

1	# Swarm Orchestration
2
3	## What This Skill Does
4
5	Orchestrates multi-agent swarms using agentic-flow's advanced coordination system. Supports mesh, hierarchical, and adaptive topologies with automatic task distribution, load balancing, and fault tolerance.
6
7	## Prerequisites
8
9	- agentic-flow v3.0.0-alpha.1+
10	- Node.js 18+
11	- Understanding of distributed systems (helpful)
12
13	## Quick Start
14
15	```bash
16	# Initialize swarm
17	npx agentic-flow hooks swarm-init --topology mesh --max-agents 5
18
19	# Spawn agents
20	npx agentic-flow hooks agent-spawn --type coder
21	npx agentic-flow hooks agent-spawn --type tester
22	npx agentic-flow hooks agent-spawn --type reviewer
23
24	# Orchestrate task
25	npx agentic-flow hooks task-orchestrate \
26	--task "Build REST API with tests" \
27	--mode parallel
28	```
29
30	## Topology Patterns
31
32	### 1. Mesh (Peer-to-Peer)
33	```typescript
34	// Equal peers, distributed decision-making
35	await swarm.init({
36	topology: 'mesh',
37	agents: ['coder', 'tester', 'reviewer'],
38	communication: 'broadcast'
39	});
40	```
41
42	### 2. Hierarchical (Queen-Worker)
43	```typescript
44	// Centralized coordination, specialized workers
45	await swarm.init({
46	topology: 'hierarchical',
47	queen: 'architect',
48	workers: ['backend-dev', 'frontend-dev', 'db-designer']
49	});
50	```
51
52	### 3. Adaptive (Dynamic)
53	```typescript
54	// Automatically switches topology based on task
55	await swarm.init({
56	topology: 'adaptive',
57	optimization: 'task-complexity'
58	});
59	```
60
61	## Task Orchestration
62
63	### Parallel Execution
64	```typescript
65	// Execute tasks concurrently
66	const results = await swarm.execute({
67	tasks: [
68	{ agent: 'coder', task: 'Implement API endpoints' },
69	{ agent: 'frontend', task: 'Build UI components' },
70	{ agent: 'tester', task: 'Write test suite' }
71	],
72	mode: 'parallel',
73	timeout: 300000 // 5 minutes
74	});
75	```
76
77	### Pipeline Execution
78	```typescript
79	// Sequential pipeline with dependencies
80	await swarm.pipeline([
81	{ stage: 'design', agent: 'architect' },
82	{ stage: 'implement', agent: 'coder', after: 'design' },
83	{ stage: 'test', agent: 'tester', after: 'implement' },
84	{ stage: 'review', agent: 'reviewer', after: 'test' }
85	]);
86	```
87
88	### Adaptive Execution
89	```typescript
90	// Let swarm decide execution strategy
91	await swarm.autoOrchestrate({
92	goal: 'Build production-ready API',
93	constraints: {
94	maxTime: 3600,
95	maxAgents: 8,
96	quality: 'high'
97	}
98	});
99	```
100
101	## Memory Coordination
102
103	```typescript
104	// Share state across swarm
105	await swarm.memory.store('api-schema', {
106	endpoints: [...],
107	models: [...]
108	});
109
110	// Agents read shared memory
111	const schema = await swarm.memory.retrieve('api-schema');
112	```
113
114	## Advanced Features
115
116	### Load Balancing
117	```typescript
118	// Automatic work distribution
119	await swarm.enableLoadBalancing({
120	strategy: 'dynamic',
121	metrics: ['cpu', 'memory', 'task-queue']
122	});
123	```
124
125	### Fault Tolerance
126	```typescript
127	// Handle agent failures
128	await swarm.setResiliency({
129	retry: { maxAttempts: 3, backoff: 'exponential' },
130	fallback: 'reassign-task'
131	});
132	```
133
134	### Performance Monitoring
135	```typescript
136	// Track swarm metrics
137	const metrics = await swarm.getMetrics();
138	// { throughput, latency, success_rate, agent_utilization }
139	```
140
141	## Integration with Hooks
142
143	```bash
144	# Pre-task coordination
145	npx agentic-flow hooks pre-task --description "Build API"
146
147	# Post-task synchronization
148	npx agentic-flow hooks post-task --task-id "task-123"
149
150	# Session restore
151	npx agentic-flow hooks session-restore --session-id "swarm-001"
152	```
153
154	## Best Practices
155
156	1. Start small: Begin with 2-3 agents, scale up
157	2. Use memory: Share context through swarm memory
158	3. Monitor metrics: Track performance and bottlenecks
159	4. Enable hooks: Automatic coordination and sync
160	5. Set timeouts: Prevent hung tasks
161
162	## Troubleshooting
163
164	### Issue: Agents not coordinating
165	Solution: Verify memory access and enable hooks
166
167	### Issue: Poor performance
168	Solution: Check topology (use adaptive) and enable load balancing
169
170	## Learn More
171
172	- Swarm Guide: docs/swarm/orchestration.md
173	- Topology Patterns: docs/swarm/topologies.md
174	- Hooks Integration: docs/hooks/coordination.md

People who liked this also viewed

6 picks

Type

swarm-orchestration

301stars24forks

View on GitHub

How to install swarm-orchestration?

spencermarx/open-code-review/swarm-orchestration

$npx -y skills add spencermarx/open-code-review --skill swarm-orchestration

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Swarm Orchestration
2
3	## What This Skill Does
4
5	Orchestrates multi-agent swarms using agentic-flow's advanced coordination system. Supports mesh, hierarchical, and adaptive topologies with automatic task distribution, load balancing, and fault tolerance.
6
7	## Prerequisites
8
9	- agentic-flow v3.0.0-alpha.1+
10	- Node.js 18+
11	- Understanding of distributed systems (helpful)
12
13	## Quick Start
14
15	```bash
16	# Initialize swarm
17	npx agentic-flow hooks swarm-init --topology mesh --max-agents 5
18
19	# Spawn agents
20	npx agentic-flow hooks agent-spawn --type coder
21	npx agentic-flow hooks agent-spawn --type tester
22	npx agentic-flow hooks agent-spawn --type reviewer
23
24	# Orchestrate task
25	npx agentic-flow hooks task-orchestrate \
26	--task "Build REST API with tests" \
27	--mode parallel
28	```
29
30	## Topology Patterns
31
32	### 1. Mesh (Peer-to-Peer)
33	```typescript
34	// Equal peers, distributed decision-making
35	await swarm.init({
36	topology: 'mesh',
37	agents: ['coder', 'tester', 'reviewer'],
38	communication: 'broadcast'
39	});
40	```
41
42	### 2. Hierarchical (Queen-Worker)
43	```typescript
44	// Centralized coordination, specialized workers
45	await swarm.init({
46	topology: 'hierarchical',
47	queen: 'architect',
48	workers: ['backend-dev', 'frontend-dev', 'db-designer']
49	});
50	```
51
52	### 3. Adaptive (Dynamic)
53	```typescript
54	// Automatically switches topology based on task
55	await swarm.init({
56	topology: 'adaptive',
57	optimization: 'task-complexity'
58	});
59	```
60
61	## Task Orchestration
62
63	### Parallel Execution
64	```typescript
65	// Execute tasks concurrently
66	const results = await swarm.execute({
67	tasks: [
68	{ agent: 'coder', task: 'Implement API endpoints' },
69	{ agent: 'frontend', task: 'Build UI components' },
70	{ agent: 'tester', task: 'Write test suite' }
71	],
72	mode: 'parallel',
73	timeout: 300000 // 5 minutes
74	});
75	```
76
77	### Pipeline Execution
78	```typescript
79	// Sequential pipeline with dependencies
80	await swarm.pipeline([
81	{ stage: 'design', agent: 'architect' },
82	{ stage: 'implement', agent: 'coder', after: 'design' },
83	{ stage: 'test', agent: 'tester', after: 'implement' },
84	{ stage: 'review', agent: 'reviewer', after: 'test' }
85	]);
86	```
87
88	### Adaptive Execution
89	```typescript
90	// Let swarm decide execution strategy
91	await swarm.autoOrchestrate({
92	goal: 'Build production-ready API',
93	constraints: {
94	maxTime: 3600,
95	maxAgents: 8,
96	quality: 'high'
97	}
98	});
99	```
100
101	## Memory Coordination
102
103	```typescript
104	// Share state across swarm
105	await swarm.memory.store('api-schema', {
106	endpoints: [...],
107	models: [...]
108	});
109
110	// Agents read shared memory
111	const schema = await swarm.memory.retrieve('api-schema');
112	```
113
114	## Advanced Features
115
116	### Load Balancing
117	```typescript
118	// Automatic work distribution
119	await swarm.enableLoadBalancing({
120	strategy: 'dynamic',
121	metrics: ['cpu', 'memory', 'task-queue']
122	});
123	```
124
125	### Fault Tolerance
126	```typescript
127	// Handle agent failures
128	await swarm.setResiliency({
129	retry: { maxAttempts: 3, backoff: 'exponential' },
130	fallback: 'reassign-task'
131	});
132	```
133
134	### Performance Monitoring
135	```typescript
136	// Track swarm metrics
137	const metrics = await swarm.getMetrics();
138	// { throughput, latency, success_rate, agent_utilization }
139	```
140
141	## Integration with Hooks
142
143	```bash
144	# Pre-task coordination
145	npx agentic-flow hooks pre-task --description "Build API"
146
147	# Post-task synchronization
148	npx agentic-flow hooks post-task --task-id "task-123"
149
150	# Session restore
151	npx agentic-flow hooks session-restore --session-id "swarm-001"
152	```
153
154	## Best Practices
155
156	1. Start small: Begin with 2-3 agents, scale up
157	2. Use memory: Share context through swarm memory
158	3. Monitor metrics: Track performance and bottlenecks
159	4. Enable hooks: Automatic coordination and sync
160	5. Set timeouts: Prevent hung tasks
161
162	## Troubleshooting
163
164	### Issue: Agents not coordinating
165	Solution: Verify memory access and enable hooks
166
167	### Issue: Poor performance
168	Solution: Check topology (use adaptive) and enable load balancing
169
170	## Learn More
171
172	- Swarm Guide: docs/swarm/orchestration.md
173	- Topology Patterns: docs/swarm/topologies.md
174	- Hooks Integration: docs/hooks/coordination.md

People who liked this also viewed

6 picks

Type