team

Name: team
Author: jmstar85

151stars12forks

How to install team?

jmstar85/oh-my-githubcopilot/team

$npx -y skills add jmstar85/oh-my-githubcopilot --skill team

Installs into the current project.

Summary

N coordinated agents on shared task list with staged pipeline. Activate when user says: team, spawn agents, parallel agents, coordinate agents, multi-agent, swarm, work together.

Files · 1

View on GitHub

SKILL.md

1	# Team
2
3	Spawn N coordinated agents working on a shared task list. Uses VS Code's native subagent system for team management, inter-agent messaging, and task dependencies.
4
5	## When to Use
6	- Task is decomposable into independent subtasks
7	- Multiple files/modules need parallel work
8	- Large-scale refactoring, migration, or multi-service work
9
10	## When NOT to Use
11	- Single-file changes → use `/omg-autopilot` or direct editing
12	- Sequential pipeline → use `/omg-autopilot`
13	- Just need a plan → use `/plan` or `/ralplan`
14
15	## Usage
16
17	```
18	/team 3:executor "fix all TypeScript errors"
19	/team 4:designer "implement responsive layouts"
20	/team "refactor the auth module"
21	/team ralph "build a complete REST API"
22	```
23
24	### Parameters
25
26	- N — Number of agents (1-20). Defaults to auto-sizing.
27	- agent-type — Agent for `team-exec` stage (executor, debugger, designer, etc.). Defaults to stage-aware routing.
28	- ralph — Wraps team in Ralph's persistence loop (retry on failure, verification before completion).
29	- task — High-level task to decompose and distribute.
30
31	## Architecture
32
33	```
34	User: "/team 3:executor fix all TypeScript errors"
35	\|
36	v
37	[omg-coordinator (Lead)]
38	\|
39	+-- Analyze & decompose → subtask list
40	\|
41	+-- Create tasks with dependencies
42	\|
43	+-- Spawn N worker agents (subagents)
44	\|
45	+-- Monitor loop (messages + task polling)
46	\|
47	+-- Completion → shutdown workers → cleanup
48	```
49
50	## Staged Pipeline
51
52	`team-plan → team-prd → team-exec → team-verify → team-fix (loop)`
53
54	### Stage Agent Routing
55
56	Each stage uses specialized agents — not just executors:
57
58	\| Stage \| Required Agents \| Optional Agents \|
59	\|-------\|----------------\|-----------------\|
60	\| team-plan \| @explore, @planner \| @analyst, @architect \|
61	\| team-prd \| @analyst \| @critic \|
62	\| team-exec \| @executor \| @debugger, @designer, @writer, @test-engineer \|
63	\| team-verify \| @verifier \| @test-engineer, @security-reviewer, @code-reviewer \|
64	\| team-fix \| @executor \| @debugger \|
65
66	Routing rules:
67	1. Lead picks agents per stage, not the user. User's `N:agent-type` only overrides `team-exec` workers.
68	2. Specialist agents complement executors. Route analysis to @architect, UI to @designer.
69	3. Security-sensitive or >20 file changes must include @security-reviewer + @code-reviewer in `team-verify`.
70
71	### Stage Entry/Exit Criteria
72
73	- team-plan: Entry = invocation. Exit = task graph ready.
74	- team-prd: Entry = scope ambiguous. Exit = acceptance criteria explicit.
75	- team-exec: Entry = tasks created + workers spawned. Exit = tasks reach terminal state.
76	- team-verify: Entry = execution pass done. Exit (pass) = gates pass. Exit (fail) = fix tasks generated.
77	- team-fix: Entry = defects found. Exit = fixes done, return to team-verify.
78
79	### Verify/Fix Loop
80
81	Continue `team-exec → team-verify → team-fix` until:
82	1. Verification passes with no required fixes, or
83	2. Max fix attempts exceeded (default: 3) → terminal `failed`
84
85	## Workflow
86
87	### Phase 1: Parse Input
88	- Extract N (agent count, validate 1-20)
89	- Extract agent-type
90	- Extract task description
91	- Check for `ralph` modifier
92
93	### Phase 2: Analyze & Decompose
94	Use @explore or @architect to analyze the codebase and break the task into N subtasks:
95	- Each subtask should be file-scoped or module-scoped
96	- Subtasks must be independent or have clear dependencies
97	- Each needs a concise subject and detailed description
98
99	### Phase 3: Create Team
100	Write team state via `omg_write_state`:
101	```json
102	{
103	"mode": "team",
104	"active": true,
105	"team_name": "fix-ts-errors",
106	"agent_count": 3,
107	"current_phase": "team-plan",
108	"task": "fix all TypeScript errors"
109	}
110	```
111
112	### Phase 4: Create Tasks
113	Create subtasks with dependencies. Pre-assign owners to avoid race conditions.
114
115	### Phase 5: Spawn Workers
116	Spawn N subagents in parallel using the `agents` field references:
117	- Each worker gets team preamble + assigned tasks
118	- Workers execute independently and report back
119
120	### Phase 6: Monitor
121	Monitor via two channels:
122	1. Inbound messages — workers report completion or need help
123	2. Task polling — check overall progress periodically
124
125	Coordination actions:
126	- Unblock a worker with guidance
127	- Reassign work if a worker finishes early
128	- Handle failures — reassign or spawn replacement
129
130	Watchdog policy:
131	- Task stuck >5 min without messages → send status check
132	- No messages >10 min → reassign task
133	- Worker fails 2+ tasks → stop assigning to it
134
135	### Phase 7: Stage Transitions
136	Update state on every stage transition via `omg_write_state`.
137
138	### Phase 8: Completion
139	1. Signal shutdown to all workers
140	2. Wait for resp

People who liked this also viewed

6 picks

Type

find-skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skSkillsJul 20262.6M27k
grill-meA relentless interview to sharpen a plan or design.SkillsJul 2026633k182k
vercel-react-best-practicesVercel's official collection of agent skillsSkillsJul 2026573k29k
improve-codebase-architectureScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.SkillsJul 2026522k182k
setup-matt-pocock-skillsConfigure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.SkillsJul 2026454k182k
triageMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.SkillsJul 2026424k182k

team

151stars12forks

View on GitHub

How to install team?

jmstar85/oh-my-githubcopilot/team

$npx -y skills add jmstar85/oh-my-githubcopilot --skill team

Installs into the current project.

Summary

N coordinated agents on shared task list with staged pipeline. Activate when user says: team, spawn agents, parallel agents, coordinate agents, multi-agent, swarm, work together.

Files · 1

View on GitHub

SKILL.md

1	# Team
2
3	Spawn N coordinated agents working on a shared task list. Uses VS Code's native subagent system for team management, inter-agent messaging, and task dependencies.
4
5	## When to Use
6	- Task is decomposable into independent subtasks
7	- Multiple files/modules need parallel work
8	- Large-scale refactoring, migration, or multi-service work
9
10	## When NOT to Use
11	- Single-file changes → use `/omg-autopilot` or direct editing
12	- Sequential pipeline → use `/omg-autopilot`
13	- Just need a plan → use `/plan` or `/ralplan`
14
15	## Usage
16
17	```
18	/team 3:executor "fix all TypeScript errors"
19	/team 4:designer "implement responsive layouts"
20	/team "refactor the auth module"
21	/team ralph "build a complete REST API"
22	```
23
24	### Parameters
25
26	- N — Number of agents (1-20). Defaults to auto-sizing.
27	- agent-type — Agent for `team-exec` stage (executor, debugger, designer, etc.). Defaults to stage-aware routing.
28	- ralph — Wraps team in Ralph's persistence loop (retry on failure, verification before completion).
29	- task — High-level task to decompose and distribute.
30
31	## Architecture
32
33	```
34	User: "/team 3:executor fix all TypeScript errors"
35	\|
36	v
37	[omg-coordinator (Lead)]
38	\|
39	+-- Analyze & decompose → subtask list
40	\|
41	+-- Create tasks with dependencies
42	\|
43	+-- Spawn N worker agents (subagents)
44	\|
45	+-- Monitor loop (messages + task polling)
46	\|
47	+-- Completion → shutdown workers → cleanup
48	```
49
50	## Staged Pipeline
51
52	`team-plan → team-prd → team-exec → team-verify → team-fix (loop)`
53
54	### Stage Agent Routing
55
56	Each stage uses specialized agents — not just executors:
57
58	\| Stage \| Required Agents \| Optional Agents \|
59	\|-------\|----------------\|-----------------\|
60	\| team-plan \| @explore, @planner \| @analyst, @architect \|
61	\| team-prd \| @analyst \| @critic \|
62	\| team-exec \| @executor \| @debugger, @designer, @writer, @test-engineer \|
63	\| team-verify \| @verifier \| @test-engineer, @security-reviewer, @code-reviewer \|
64	\| team-fix \| @executor \| @debugger \|
65
66	Routing rules:
67	1. Lead picks agents per stage, not the user. User's `N:agent-type` only overrides `team-exec` workers.
68	2. Specialist agents complement executors. Route analysis to @architect, UI to @designer.
69	3. Security-sensitive or >20 file changes must include @security-reviewer + @code-reviewer in `team-verify`.
70
71	### Stage Entry/Exit Criteria
72
73	- team-plan: Entry = invocation. Exit = task graph ready.
74	- team-prd: Entry = scope ambiguous. Exit = acceptance criteria explicit.
75	- team-exec: Entry = tasks created + workers spawned. Exit = tasks reach terminal state.
76	- team-verify: Entry = execution pass done. Exit (pass) = gates pass. Exit (fail) = fix tasks generated.
77	- team-fix: Entry = defects found. Exit = fixes done, return to team-verify.
78
79	### Verify/Fix Loop
80
81	Continue `team-exec → team-verify → team-fix` until:
82	1. Verification passes with no required fixes, or
83	2. Max fix attempts exceeded (default: 3) → terminal `failed`
84
85	## Workflow
86
87	### Phase 1: Parse Input
88	- Extract N (agent count, validate 1-20)
89	- Extract agent-type
90	- Extract task description
91	- Check for `ralph` modifier
92
93	### Phase 2: Analyze & Decompose
94	Use @explore or @architect to analyze the codebase and break the task into N subtasks:
95	- Each subtask should be file-scoped or module-scoped
96	- Subtasks must be independent or have clear dependencies
97	- Each needs a concise subject and detailed description
98
99	### Phase 3: Create Team
100	Write team state via `omg_write_state`:
101	```json
102	{
103	"mode": "team",
104	"active": true,
105	"team_name": "fix-ts-errors",
106	"agent_count": 3,
107	"current_phase": "team-plan",
108	"task": "fix all TypeScript errors"
109	}
110	```
111
112	### Phase 4: Create Tasks
113	Create subtasks with dependencies. Pre-assign owners to avoid race conditions.
114
115	### Phase 5: Spawn Workers
116	Spawn N subagents in parallel using the `agents` field references:
117	- Each worker gets team preamble + assigned tasks
118	- Workers execute independently and report back
119
120	### Phase 6: Monitor
121	Monitor via two channels:
122	1. Inbound messages — workers report completion or need help
123	2. Task polling — check overall progress periodically
124
125	Coordination actions:
126	- Unblock a worker with guidance
127	- Reassign work if a worker finishes early
128	- Handle failures — reassign or spawn replacement
129
130	Watchdog policy:
131	- Task stuck >5 min without messages → send status check
132	- No messages >10 min → reassign task
133	- Worker fails 2+ tasks → stop assigning to it
134
135	### Phase 7: Stage Transitions
136	Update state on every stage transition via `omg_write_state`.
137
138	### Phase 8: Completion
139	1. Signal shutdown to all workers
140	2. Wait for resp

People who liked this also viewed

6 picks

Type