agent-teams
Skill0 starsChanged 6 months agoMITView on GitHub
Use this skill when coordinating parallel work, creating agent teams, or when the task would benefit from multiple specialists working simultaneously. Triggers: 'agent team', 'parallel', 'spawn teammates', 'create a team', 'work in parallel', 'multi-agent', 'review from multiple angles', 'competing hypotheses', 'split the work', 'team of agents', or when a task naturally decomposes into independent parallel workstreams like full-stack features, code reviews, security audits, debugging investigations, refactors, or test coverage sprints.
16 other projects use a version of this.
Adopted by
Most recently taken up by
- jonmumm/skillstheir own version, first seen Oct 21, 2026
- yasunori0418/skillstheir own version, first seen Oct 13, 2026
- cubetribe/ClaudeCode_GodMode-Ontheir own version, first seen Oct 13, 2026
- sanad-source/agent-teamstheir own version, first seen Oct 13, 2026
- alinaqi/claude-bootstraptheir own version, first seen Oct 8, 2026
- ProxiBlue/claude-skillstheir own version, first seen Oct 7, 2026
Used with
Projects with it also use codex-review and security-audit.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
juriwiensSkillcodex-review
Run code review via Codex CLI
Also in 3 projects that use this
CloudflareSkillsecurity-audit
Security guidance and vulnerability review for codebases, APIs, services, CLI tools, libraries, and daemons. Use for security questions, focused reviews, vulnerability research, security audits, or pen tests. Run the complete workflow only for explicit codebase audit or pen-test requests, full/comprehensive/end-to-end reviews, or requested report artifacts.
Also in 3 projects that use this
Other versions16 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 16
- rohitg00/pro-workflow★ 2.9ksame name
- alinaqi/maggy★ 711same name
- alinaqi/claude-bootstrap★ 708same name
- travisjneuman/.claude★ 106same name
- cubetribe/ClaudeCode_GodMode-On★ 55same name
- ProxiBlue/claude-skills★ 18same name
- Abhipaddy8/outreach-agent★ 17same name
- TurboKach/claude-code-setup★ 11same name
- sergeeey/Claude-cod-top-2026★ 9same name
- yasunori0418/skills★ 9same name
6 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository55 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- agent-developmentSkill
- ai-sdkSkill
- analyzing-financial-statementsSkill
- cache-componentsSkill
- chrome-devtoolsSkill
- Codex-automation-recommenderSkill
- Codex-md-improverSkill
- code-reviewSkill
- Command DevelopmentSkill
- creating-financial-modelsSkill
- data-visualizationSkill
- debugSkill
- develop-ai-functions-exampleSkill
- evolve-ai-layerSkill
- excalidraw-diagramSkill
- find-skillsSkill
- frontend-designSkill
- Hook DevelopmentSkill
- make-skill-templateSkill
- markdown-to-htmlSkill
- mcp-builderSkill
- MCP IntegrationSkill
- pdfSkill
- playwrightSkill
- Plugin SettingsSkill
- Plugin StructureSkill
- refactorSkill
- researchSkill
- security-best-practicesSkill
- skill-creatorSkill
- Skill DevelopmentSkill
- smithery-ai-cliSkill
- source-command-commit-and-deploySkill
- source-command-commitSkill
- source-command-create-prdSkill
- source-command-create-rulesSkill
- source-command-empty-redeploySkill
- source-command-evaluateSkill
- source-command-executeSkill
- source-command-handoffSkill
- source-command-plan-featureSkill
- source-command-primeSkill
- source-command-reviewSkill
- source-command-self-healSkill
- source-command-validateSkill
- sql-queriesSkill
- statistical-analysisSkill
- vscode-ext-localizationSkill
- webapp-testingSkill
- web-artifacts-builderSkill
- web-design-guidelinesSkill
- Writing Hookify RulesSkill
- xlsxSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 13 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use agent teams when parallel exploration or implementation adds real value.
- Distinctive: line 93
Ask the lead to "shut down the team" when done
- Distinctive: line 94
If the lead starts doing work itself: "Wait for your teammates to complete their tasks"
“Common” is measured: kept by at least 6 in 10 of the 13 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Agent Teams: Patterns and Prompts
Use agent teams when parallel exploration or implementation adds real value. Each teammate is a separate Codex instance with its own context window. They communicate directly with each other and coordinate through a shared task list.
When to Use Agent Teams vs Sub-Agents
- Agent teams: teammates need to communicate, challenge each other, or coordinate on shared artifacts
- Sub-agents: focused tasks where only the result matters (research, exploration, analysis)
- Single session: sequential work, quick changes, or tasks with heavy dependencies between steps
Key Rules
- 3-5 teammates is the sweet spot — more adds overhead without proportional benefit
- Each teammate owns different files — two editing the same file causes overwrites
- Tell the lead to wait — otherwise it starts implementing instead of delegating
- Use plan approval for risky tasks — "Require plan approval before making changes"
- Give context in the prompt — teammates don't inherit your conversation history
Prompt Template
Create an agent team to [OBJECTIVE]. Spawn [N] teammates:
1. [Role] — [responsibilities, files they own, what they do first]
2. [Role] — [responsibilities, files they own, what they do first]
3. [Role] — [responsibilities, files they own, what they do first]
Coordination rules:
- [Who goes first / dependencies]
- [How they share findings]
- [File ownership — no overlap]
- [Validation to run at the end]
Wait for all teammates to finish before synthesizing results.Ready-to-Use Patterns
Code Review (3 teammates)
Spawn a security reviewer, performance reviewer, and correctness reviewer. Each applies a different lens to the same changes. They share findings and challenge each other. Synthesize a prioritized final review.
Full-Stack Feature (3 teammates)
Spawn a backend developer, frontend developer, and test engineer. Backend and frontend agree on API contract first. Test engineer writes tests as the others build. No teammate edits files owned by another.
Competing Hypotheses Debug (4 teammates)
Spawn 4 investigators, each with a different theory about the bug. They actively try to disprove each other's hypotheses. The surviving theory with the strongest evidence wins.
Architecture Review (4 teammates)
Spawn a dependency analyst, complexity assessor, pattern consistency checker, and scalability reviewer. Each examines the codebase from a different angle. Cross-reference findings — coupling issues often explain complexity hotspots.
Security Audit (4 teammates)
Spawn a secrets scanner, injection auditor, auth/authz auditor, and dependency auditor. Each focuses on a different attack surface. They challenge each other's findings to reduce false positives.
Pre-Release Test Blitz (5 teammates)
Spawn happy path tester, error path tester, boundary tester, auth tester, and data integrity tester. Each works independently on different test categories. Share bugs found so others can check for related issues.
Large-Scale Refactor (4 teammates)
Spawn a planner, two implementers (each owning different files), and a validator who runs tests continuously. Planner creates the plan first, gets approval, then implementers execute in parallel.
Performance Audit (4 teammates)
Spawn frontend, API, database, and infrastructure performance analysts. Each profiles their layer. Share findings that cross boundaries — a slow API often traces to a missing DB index.
Documentation Overhaul (4 teammates)
Spawn a doc auditor, README writer, API documenter, and rules updater. Auditor goes first to identify problems. The other three fix their respective areas in parallel.
Post-Incident Analysis (4 teammates)
Spawn a timeline builder, root cause analyst, blast radius assessor, and prevention planner. They share findings throughout. Prevention planner challenges: "If we fix that, could this still happen differently?"
Technical Breakdown for Planning (3 teammates)
Spawn a requirements analyst, architecture planner, and effort estimator. Work sequentially: requirements first, then architecture, then estimates. Each challenges the previous teammate's work.
Managing Teams
Shift+Downto cycle between teammates- Talk to any teammate directly by cycling to them and typing
Ctrl+Tto toggle the task list- Ask the lead to "shut down the team" when done
- If the lead starts doing work itself: "Wait for your teammates to complete their tasks"
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

