Agent Teams - CK-Native Orchestration Engine

Coordinate multiple independent Codex sessions. Each teammate has own context window, loads project context (AGENTS.md, skills, agents), communicates via shared task list and messaging.

Requires: Agent Teams enabled. Set CODEX_EXPERIMENTAL_AGENT_TEAMS=1 in settings.json env if using Codex < 2.1.33. May be GA in 2.1.33+.

Usage

$team <template> <context> [flags]

Templates: ck:research, ck:cook, ck:code-review, ck:debug

Flags:

• --devs N | --researchers N | --reviewers N | --debuggers N — team size
• --plan-approval / --no-plan-approval — plan gate (default: on for cook)
• --delegate — lead only coordinates, never touches code

Execution Protocol

Pre-flight (MANDATORY — merged into step 2 of every template):

1. Step 2 of every template calls TeamCreate(team_name: "...", ...). Do NOT check whether the tool exists first — just call it.
2. If the call SUCCEEDS: continue with the template.
3. If the call returns an ERROR or is unrecognized: STOP. Tell user: "Agent Teams requires CODEX_EXPERIMENTAL_AGENT_TEAMS=1 in settings.json. Team mode is not available."
4. Do NOT fall back to subagents. $team MUST use Agent Teams or abort.
5. All teammate spawns MUST include team_name parameter — never spawn Task without it.

When activated, IMMEDIATELY execute the matching template sequence below. Do NOT ask for confirmation. Do NOT explain what you're about to do. Execute the tool calls in order. Report progress after each major step.

---

CK Context Block

Every teammate spawn prompt MUST include this context at the end:

CK Context:
- Work dir: {CK_PROJECT_ROOT or CWD}
- Reports: {CK_REPORTS_PATH or "plans/reports/"}
- Plans: {CK_PLANS_PATH or "plans/"}
- Branch: {CK_GIT_BRANCH or current branch}
- Naming: {CK_NAME_PATTERN or "YYMMDD-HHMM"}
- Active plan: {CK_ACTIVE_PLAN or "none"}
- Commits: conventional (feat:, fix:, docs:, refactor:, test:, chore:)
- Refer to teammates by NAME, not agent ID

---

ON $team research <topic> [--researchers N]:

Wraps $research skill — scope, gather, analyze, report.

IMMEDIATELY execute in order:

1. Derive N angles from <topic> (default N=3):

   • Angle 1: Architecture, patterns, proven approaches
   • Angle 2: Alternatives, competing solutions, trade-offs
   • Angle 3: Risks, edge cases, failure modes, security
   • (If N>3, derive additional angles from topic context)

2. CALL TeamCreate(team_name: "<topic-slug>")

3. CALL TaskCreate x N — one per angle:

   • Subject: Research: <angle-title>
   • Description: Investigate <angle> for topic: <topic>. Save report to: {CK_REPORTS_PATH}/researcher-{N}-{CK_NAME_PATTERN}-{topic-slug}.md. Format: Executive summary, key findings, evidence, recommendations. Mark task completed when done. Send findings summary to lead.

4. CALL Task x N to spawn researcher teammates:

   • subagent_type: "researcher", team_name: "<topic-slug>", model: "haiku"
   • name: "researcher-{N}"
   • Prompt: task description + CK Context Block

5. MONITOR via TaskCompleted hook events + TaskList fallback:

   • TaskCompleted events auto-notify when researchers finish
   • Fallback: Check TaskList if no event received in 60s
   • If stuck >5 min, message teammate directly

6. READ all researcher reports from {CK_REPORTS_PATH}/

7. SYNTHESIZE into: {CK_REPORTS_PATH}/research-summary-{CK_NAME_PATTERN}-{topic-slug}.md Format: exec summary, key findings, comparative analysis, recommendations, unresolved questions.

8. SHUTDOWN: SendMessage(type: "shutdown_request") to each teammate

9. CLEANUP: TeamDelete (no parameters — just call it)

10. REPORT: Tell user Research complete. Summary: {path}. N reports generated.

---

ON $team cook <plan-path-or-description> [--devs N]:

Wraps ck:cook skill — plan, code, test, review, finalize.

IMMEDIATELY execute in order:

1. READ plan (if path provided) OR create via planner teammate:

   • If description only: spawn Task(subagent_type: "planner") to create plan first
   • Parse plan into N independent task groups with file ownership boundaries

2. CALL TeamCreate(team_name: "<feature-slug>")

3. CALL TaskCreate x (N + 1) — N dev tasks + 1 tester task:

   • Dev tasks: include File ownership: <glob patterns> — NO overlap between devs
   • Tester task: addBlockedBy all dev task IDs
   • Each task description includes: implementation scope, file ownership, acceptance criteria

4. CALL Task x N to spawn developer teammates:

   • subagent_type: "fullstack_developer", mode: "plan"
   • model: "sonnet", name: "dev-{N}"
   • Prompt: task description + CK Context Block
   • REVIEW and APPROVE each developer's plan via plan_approval_response

5. MONITOR dev completion via TaskCompleted events:

   • TaskCompleted hook notifies when each dev task finishes
   • When all N dev tasks show completed, spawn tester immediately
   • TeammateIdle events confirm devs are available for shutdown
   • Fallback: Check TaskList if no events received in 60s
   • Task(subagent_type: "tester", model: "haiku", name: "tester")
   • Tester runs full test suite, reports pass/fail

6. DOCS SYNC EVAL (MANDATORY for cook — from ck:cook finalize):

   Docs impact: [none|minor|major]
   Action: [no update needed — <reason>] | [updated <page>] | [needs separate PR]
   

7. SHUTDOWN all teammates via SendMessage(type: "shutdown_request")

8. CLEANUP: TeamDelete (no parameters — just call it)

9. REPORT: Tell user what was cooked, test results, docs impact.

---

ON $team review <scope> [--reviewers N]:

Wraps $code-review skill — scout, review, synthesize with evidence gates.

IMMEDIATELY execute in order:

1. DERIVE N review focuses from <scope> (default N=3):

   • Focus 1: Security — vulnerabilities, auth, input validation, OWASP
   • Focus 2: Performance — bottlenecks, memory, complexity, scaling
   • Focus 3: Test coverage — gaps, edge cases, error paths
   • (If N>3, derive from scope: architecture, DX, accessibility, etc.)

2. CALL TeamCreate(team_name: "review-<scope-slug>")

3. CALL TaskCreate x N — one per focus:

   • Subject: Review: <focus-title>
   • Description: Review <scope> for <focus>. Output severity-rated findings only. Format: [CRITICAL|IMPORTANT|MODERATE] <finding> — <evidence> — <recommendation>. No "seems" or "probably" — concrete evidence only. Save to: {CK_REPORTS_PATH}/reviewer-{N}-{CK_NAME_PATTERN}-{scope-slug}.md. Mark task completed when done.

4. CALL Task x N to spawn reviewers:

   • subagent_type: "code_reviewer", model: "haiku", name: "reviewer-{N}"
   • Prompt: task description + CK Context Block

5. MONITOR via TaskCompleted hook events + TaskList fallback:

   • TaskCompleted events auto-notify when reviewers finish
   • Fallback: Check TaskList if no event received in 60s

6. SYNTHESIZE into: {CK_REPORTS_PATH}/review-{scope-slug}.md

   • Deduplicate findings across reviewers
   • Prioritize by severity: CRITICAL > IMPORTANT > MODERATE
   • Create action items list with owners

7. SHUTDOWN all teammates via SendMessage(type: "shutdown_request")

8. CLEANUP: TeamDelete (no parameters — just call it)

9. REPORT: Tell user Review complete. {X} findings ({Y} critical). Report: {path}.

---

ON $team debug <issue> [--debuggers N]:

Wraps $fix skill — root-cause-first, adversarial hypotheses, disprove to converge.

IMMEDIATELY execute in order:

1. GENERATE N competing hypotheses from <issue> (default N=3):

   • Each hypothesis must be independently testable
   • Each must predict different observable symptoms
   • Frame as: "If <cause>, then we should see <evidence>"

2. CALL TeamCreate(team_name: "debug-<issue-slug>")

3. CALL TaskCreate x N — one per hypothesis:

   • Subject: Debug: Test hypothesis — <theory>
   • Description: Investigate hypothesis: <theory>. For issue: <issue>. ADVERSARIAL: actively try to disprove other theories. Message other debuggers to challenge findings. Report evidence FOR and AGAINST your theory. Save findings to: {CK_REPORTS_PATH}/debugger-{N}-{CK_NAME_PATTERN}-{issue-slug}.md. Mark task completed when done.

4. CALL Task x N to spawn debugger teammates:

   • subagent_type: "debugger", model: "sonnet", name: "debugger-{N}"
   • Prompt: task description + CK Context Block

5. MONITOR via TaskCompleted events. Debuggers should message each other — let them converge.

   • TaskCompleted events notify as each hypothesis is tested
   • TeammateIdle events indicate debugger awaiting peer input
   • Fallback: Check TaskList if no events in 60s

6. READ all debugger reports. Identify surviving theory as root cause.

7. WRITE root cause report: {CK_REPORTS_PATH}/debug-{issue-slug}.md Format: Root cause, evidence chain, disproven hypotheses, recommended fix.

8. SHUTDOWN all teammates via SendMessage(type: "shutdown_request")

9. CLEANUP: TeamDelete (no parameters — just call it)

10. REPORT: Tell user Debug complete. Root cause: <summary>. Report: {path}.

---

Agent Memory

Agents with memory: project retain learnings across team sessions. Memory persists in .codex/agent-memory/<name>/ (gitignored). Useful for:

• Code reviewer remembering project conventions
• Debugger recalling past failure patterns
• Tester tracking flaky tests and coverage gaps

---

Agent Memory

Agents with memory: project retain learnings across team sessions. Memory persists in .codex/agent-memory/<name>/ (gitignored). Useful for:

• Code reviewer remembering project conventions
• Debugger recalling past failure patterns
• Tester tracking flaky tests and coverage gaps

---

When to Use Agent Teams vs Subagents

Scenario	Subagents (Task tool)	Agent Teams
Focused task (test, lint, single review)	Yes	Overkill
Sequential chain (plan → code → test)	Yes	No
3+ independent parallel workstreams	Maybe	Yes
Competing debug hypotheses	No	Yes
Cross-layer work (FE + BE + tests)	Maybe	Yes
Workers need to discuss/challenge findings	No	Yes
Token budget is tight	Yes	No (high cost)

Token Budget

Template	Estimated Tokens	Model Strategy
Research (3)	~150K-300K	haiku for all
Cook (4)	~400K-800K	sonnet for devs, haiku for tester
Review (3)	~100K-200K	haiku for all
Debug (3)	~200K-400K	sonnet for all

Error Recovery

1. Check status: Shift+Up/Down (in-process) or click pane (split)
2. Redirect: Send direct message with corrective instructions
3. Replace: Shut down failed teammate, spawn replacement for same task
4. Reassign: TaskUpdate stuck task to unblock dependents

Abort Team

Shut down all teammates. Then call TeamDelete (no parameters).

If unresponsive: close terminal or kill session. Clean orphaned configs at ~/.codex/teams/ manually.

Display Modes

• auto (default): split panes if in tmux, otherwise in-process
• in-process: all in one terminal. Shift+Up/Down navigate. Ctrl+T task list.
• tmux/split: each teammate own pane. Requires tmux or iTerm2.

Rules Reference

See .codex/rules/team-coordination-rules.md for teammate behavior rules.

v2.1.0: Event-driven orchestration via TaskCompleted/TeammateIdle hooks. Agent memory + Task restrictions.