parallel-agents

Name: parallel-agents
Author: One-Man-Company

77stars16forks

How to install parallel-agents?

one-man-company/skills-contextmanager/parallel-agents

$npx -y skills add One-Man-Company/Skills-ContextManager --skill parallel-agents

Installs into the current project.

Summary

Multi-agent orchestration patterns. Use when multiple independent tasks can run with different domain expertise or when comprehensive analysis requires multiple perspectives.

Files · 1

View on GitHub

SKILL.md

1	# Native Parallel Agents
2
3	> Orchestration through Antigravity's built-in Agent Tool
4
5	## Overview
6
7	This skill enables coordinating multiple specialized agents through Antigravity's native agent system. Unlike external scripts, this approach keeps all orchestration within Antigravity's control.
8
9	## When to Use Orchestration
10
11	✅ Good for:
12	- Complex tasks requiring multiple expertise domains
13	- Code analysis from security, performance, and quality perspectives
14	- Comprehensive reviews (architecture + security + testing)
15	- Feature implementation needing backend + frontend + database work
16
17	❌ Not for:
18	- Simple, single-domain tasks
19	- Quick fixes or small changes
20	- Tasks where one agent suffices
21
22	---
23
24	## Native Agent Invocation
25
26	### Single Agent
27	```
28	Use the security-auditor agent to review authentication
29	```
30
31	### Sequential Chain
32	```
33	First, use the explorer-agent to discover project structure.
34	Then, use the backend-specialist to review API endpoints.
35	Finally, use the test-engineer to identify test gaps.
36	```
37
38	### With Context Passing
39	```
40	Use the frontend-specialist to analyze React components.
41	Based on those findings, have the test-engineer generate component tests.
42	```
43
44	### Resume Previous Work
45	```
46	Resume agent [agentId] and continue with additional requirements.
47	```
48
49	---
50
51	## Orchestration Patterns
52
53	### Pattern 1: Comprehensive Analysis
54	```
55	Agents: explorer-agent → [domain-agents] → synthesis
56
57	1. explorer-agent: Map codebase structure
58	2. security-auditor: Security posture
59	3. backend-specialist: API quality
60	4. frontend-specialist: UI/UX patterns
61	5. test-engineer: Test coverage
62	6. Synthesize all findings
63	```
64
65	### Pattern 2: Feature Review
66	```
67	Agents: affected-domain-agents → test-engineer
68
69	1. Identify affected domains (backend? frontend? both?)
70	2. Invoke relevant domain agents
71	3. test-engineer verifies changes
72	4. Synthesize recommendations
73	```
74
75	### Pattern 3: Security Audit
76	```
77	Agents: security-auditor → penetration-tester → synthesis
78
79	1. security-auditor: Configuration and code review
80	2. penetration-tester: Active vulnerability testing
81	3. Synthesize with prioritized remediation
82	```
83
84	---
85
86	## Available Agents
87
88	\| Agent \| Expertise \| Trigger Phrases \|
89	\|-------\|-----------\|-----------------\|
90	\| `orchestrator` \| Coordination \| "comprehensive", "multi-perspective" \|
91	\| `security-auditor` \| Security \| "security", "auth", "vulnerabilities" \|
92	\| `penetration-tester` \| Security Testing \| "pentest", "red team", "exploit" \|
93	\| `backend-specialist` \| Backend \| "API", "server", "Node.js", "Express" \|
94	\| `frontend-specialist` \| Frontend \| "React", "UI", "components", "Next.js" \|
95	\| `test-engineer` \| Testing \| "tests", "coverage", "TDD" \|
96	\| `devops-engineer` \| DevOps \| "deploy", "CI/CD", "infrastructure" \|
97	\| `database-architect` \| Database \| "schema", "Prisma", "migrations" \|
98	\| `mobile-developer` \| Mobile \| "React Native", "Flutter", "mobile" \|
99	\| `api-designer` \| API Design \| "REST", "GraphQL", "OpenAPI" \|
100	\| `debugger` \| Debugging \| "bug", "error", "not working" \|
101	\| `explorer-agent` \| Discovery \| "explore", "map", "structure" \|
102	\| `documentation-writer` \| Documentation \| "write docs", "create README", "generate API docs" \|
103	\| `performance-optimizer` \| Performance \| "slow", "optimize", "profiling" \|
104	\| `project-planner` \| Planning \| "plan", "roadmap", "milestones" \|
105	\| `seo-specialist` \| SEO \| "SEO", "meta tags", "search ranking" \|
106	\| `game-developer` \| Game Development \| "game", "Unity", "Godot", "Phaser" \|
107
108	---
109
110	## Antigravity Built-in Agents
111
112	These work alongside custom agents:
113
114	\| Agent \| Model \| Purpose \|
115	\|-------\|-------\|---------\|
116	\| Explore \| Haiku \| Fast read-only codebase search \|
117	\| Plan \| Sonnet \| Research during plan mode \|
118	\| General-purpose \| Sonnet \| Complex multi-step modifications \|
119
120	Use Explore for quick searches, custom agents for domain expertise.
121
122	---
123
124	## Synthesis Protocol
125
126	After all agents complete, synthesize:
127
128	```markdown
129	## Orchestration Synthesis
130
131	### Task Summary
132	[What was accomplished]
133
134	### Agent Contributions
135	\| Agent \| Finding \|
136	\|-------\|---------\|
137	\| security-auditor \| Found X \|
138	\| backend-specialist \| Identified Y \|
139
140	### Consolidated Recommendations
141	1. Critical: [Issue from Agent A]
142	2. Important: [Issue from Agent B]
143	3. Nice-to-have: [Enhancement from Agent C]
144
145	### Action Items
146	- [ ] Fix critical security issue
147	- [ ] Refactor API endpoint
148	- [ ] Add missing tests
149	```
150
151	---
152
153	## Best Practices
154
155	1. Available agents - 17 specialized agents can be orchestrated
156	2. Logical order - Discovery → Analysis → Implementation → Testing
157	3. Share context - Pass relevant findings to subsequent agents
158	4. Single synthesis - One unified report, not separate outputs
159	5. Verify changes - Always include test-engineer for code modifications
160
161	--

People who liked this also viewed

6 picks

Type

parallel-agents

77stars16forks

View on GitHub

How to install parallel-agents?

one-man-company/skills-contextmanager/parallel-agents

$npx -y skills add One-Man-Company/Skills-ContextManager --skill parallel-agents

Installs into the current project.

Summary

Multi-agent orchestration patterns. Use when multiple independent tasks can run with different domain expertise or when comprehensive analysis requires multiple perspectives.

Files · 1

View on GitHub

SKILL.md

1	# Native Parallel Agents
2
3	> Orchestration through Antigravity's built-in Agent Tool
4
5	## Overview
6
7	This skill enables coordinating multiple specialized agents through Antigravity's native agent system. Unlike external scripts, this approach keeps all orchestration within Antigravity's control.
8
9	## When to Use Orchestration
10
11	✅ Good for:
12	- Complex tasks requiring multiple expertise domains
13	- Code analysis from security, performance, and quality perspectives
14	- Comprehensive reviews (architecture + security + testing)
15	- Feature implementation needing backend + frontend + database work
16
17	❌ Not for:
18	- Simple, single-domain tasks
19	- Quick fixes or small changes
20	- Tasks where one agent suffices
21
22	---
23
24	## Native Agent Invocation
25
26	### Single Agent
27	```
28	Use the security-auditor agent to review authentication
29	```
30
31	### Sequential Chain
32	```
33	First, use the explorer-agent to discover project structure.
34	Then, use the backend-specialist to review API endpoints.
35	Finally, use the test-engineer to identify test gaps.
36	```
37
38	### With Context Passing
39	```
40	Use the frontend-specialist to analyze React components.
41	Based on those findings, have the test-engineer generate component tests.
42	```
43
44	### Resume Previous Work
45	```
46	Resume agent [agentId] and continue with additional requirements.
47	```
48
49	---
50
51	## Orchestration Patterns
52
53	### Pattern 1: Comprehensive Analysis
54	```
55	Agents: explorer-agent → [domain-agents] → synthesis
56
57	1. explorer-agent: Map codebase structure
58	2. security-auditor: Security posture
59	3. backend-specialist: API quality
60	4. frontend-specialist: UI/UX patterns
61	5. test-engineer: Test coverage
62	6. Synthesize all findings
63	```
64
65	### Pattern 2: Feature Review
66	```
67	Agents: affected-domain-agents → test-engineer
68
69	1. Identify affected domains (backend? frontend? both?)
70	2. Invoke relevant domain agents
71	3. test-engineer verifies changes
72	4. Synthesize recommendations
73	```
74
75	### Pattern 3: Security Audit
76	```
77	Agents: security-auditor → penetration-tester → synthesis
78
79	1. security-auditor: Configuration and code review
80	2. penetration-tester: Active vulnerability testing
81	3. Synthesize with prioritized remediation
82	```
83
84	---
85
86	## Available Agents
87
88	\| Agent \| Expertise \| Trigger Phrases \|
89	\|-------\|-----------\|-----------------\|
90	\| `orchestrator` \| Coordination \| "comprehensive", "multi-perspective" \|
91	\| `security-auditor` \| Security \| "security", "auth", "vulnerabilities" \|
92	\| `penetration-tester` \| Security Testing \| "pentest", "red team", "exploit" \|
93	\| `backend-specialist` \| Backend \| "API", "server", "Node.js", "Express" \|
94	\| `frontend-specialist` \| Frontend \| "React", "UI", "components", "Next.js" \|
95	\| `test-engineer` \| Testing \| "tests", "coverage", "TDD" \|
96	\| `devops-engineer` \| DevOps \| "deploy", "CI/CD", "infrastructure" \|
97	\| `database-architect` \| Database \| "schema", "Prisma", "migrations" \|
98	\| `mobile-developer` \| Mobile \| "React Native", "Flutter", "mobile" \|
99	\| `api-designer` \| API Design \| "REST", "GraphQL", "OpenAPI" \|
100	\| `debugger` \| Debugging \| "bug", "error", "not working" \|
101	\| `explorer-agent` \| Discovery \| "explore", "map", "structure" \|
102	\| `documentation-writer` \| Documentation \| "write docs", "create README", "generate API docs" \|
103	\| `performance-optimizer` \| Performance \| "slow", "optimize", "profiling" \|
104	\| `project-planner` \| Planning \| "plan", "roadmap", "milestones" \|
105	\| `seo-specialist` \| SEO \| "SEO", "meta tags", "search ranking" \|
106	\| `game-developer` \| Game Development \| "game", "Unity", "Godot", "Phaser" \|
107
108	---
109
110	## Antigravity Built-in Agents
111
112	These work alongside custom agents:
113
114	\| Agent \| Model \| Purpose \|
115	\|-------\|-------\|---------\|
116	\| Explore \| Haiku \| Fast read-only codebase search \|
117	\| Plan \| Sonnet \| Research during plan mode \|
118	\| General-purpose \| Sonnet \| Complex multi-step modifications \|
119
120	Use Explore for quick searches, custom agents for domain expertise.
121
122	---
123
124	## Synthesis Protocol
125
126	After all agents complete, synthesize:
127
128	```markdown
129	## Orchestration Synthesis
130
131	### Task Summary
132	[What was accomplished]
133
134	### Agent Contributions
135	\| Agent \| Finding \|
136	\|-------\|---------\|
137	\| security-auditor \| Found X \|
138	\| backend-specialist \| Identified Y \|
139
140	### Consolidated Recommendations
141	1. Critical: [Issue from Agent A]
142	2. Important: [Issue from Agent B]
143	3. Nice-to-have: [Enhancement from Agent C]
144
145	### Action Items
146	- [ ] Fix critical security issue
147	- [ ] Refactor API endpoint
148	- [ ] Add missing tests
149	```
150
151	---
152
153	## Best Practices
154
155	1. Available agents - 17 specialized agents can be orchestrated
156	2. Logical order - Discovery → Analysis → Implementation → Testing
157	3. Share context - Pass relevant findings to subsequent agents
158	4. Single synthesis - One unified report, not separate outputs
159	5. Verify changes - Always include test-engineer for code modifications
160
161	--

People who liked this also viewed

6 picks

Type