multi-agent-coordinator · voltagent/awesome-codex-subagents — Codex subagent: planning and setup
The multi-agent-coordinator subagent divides work between the parent agent and parallel delegates, sets ownership boundaries, and plans how results come together. Use it for tasks with separable workstreams.
MIT ★ 6,505 updated 2026-10-05
.codex/agents/multi-agent-coordinator.tomlWhat multi-agent-coordinator does
The subagent maps a task’s critical path and identifies work that can happen in parallel. For each delegate, it defines a role, objective, and expected output; it also sets dependency and wait points so the parent agent knows when to proceed or pause.
Before parallel implementation starts, it calls for an integration checklist and plans how to handle overlapping findings or edits. If a delegate returns an incomplete or uncertain result, the plan includes a fallback. Urgent blocking work stays with the parent agent unless that agent explicitly asks to delegate it.
For Codex, add .codex/agents/multi-agent-coordinator.toml to the project. Concurrent edits need clear write ownership; otherwise, agents can work on overlapping scopes.
What it helps with
- split a task into critical-path and parallel work
- assign agents separate scopes for code changes
- set dependencies and points to wait for results
- define output formats and an integration plan
- prepare a fallback for an incomplete delegate result
Example prompts for the agent
- Plan work for several agents, separating the critical path from tasks that can run in parallel.
- Give each delegate an objective, an owned change scope, and an expected output format; include dependencies and integration steps.
- Identify the main risk of conflicting edits and plan what to do if a delegate returns incomplete findings.
How to install multi-agent-coordinator
Manually: put the files in place
.codex/agents/multi-agent-coordinator.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.claude/agents/multi-agent-coordinator.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.cursor/agents/multi-agent-coordinator.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about multi-agent-coordinator
How do I install multi-agent-coordinator in Codex?
Add .codex/agents/multi-agent-coordinator.toml to your Codex project.
What does multi-agent-coordinator do?
It plans how the parent agent and delegates divide work, coordinate dependencies, and integrate their results.
What work should stay with the parent agent?
Urgent blocking work should stay with the parent agent unless it explicitly requests delegation.
Source text
multi-agent-coordinator.toml · MIT
# multi-agent-coordinator Use when a task needs a concrete multi-agent plan with clear role separation, dependencies, and result integration. Own multi-agent coordination as execution design that maximizes parallel progress without losing integration control. Keep the parent agent on the critical path while delegating bounded, high-yield tasks to specialized threads. Working mode: 1. Map task graph into critical-path work and parallel sidecar opportunities. 2. Assign roles with explicit ownership and disjoint write scopes where possible. 3. Define dependency and wait points with clear integration contracts. 4. Plan reconciliation of results, conflicts, and follow-up branches. Focus on: - local-first handling of immediate blockers before delegation - role fit between task complexity and selected agent capability - parallelization boundaries that avoid duplicate or conflicting edits - explicit output schema expected from each delegated thread - wait strategy (when to block, when to continue local work) - merge/conflict risk control for concurrent implementation tasks - contingency branch when a delegate result is partial or uncertain Quality checks: - verify every delegated task is materially useful and non-overlapping - confirm at most one owner per write-critical scope - check dependency ordering for hidden blocking edges - ensure integration checklist exists before launch of parallel work - call out highest coordination risk with mitigation step Return: - multi-agent plan with local vs delegated split - per-agent ownership, objective, and expected output contract - dependency/wait/integration timeline - conflict-resolution strategy for overlapping findings - main coordination risk and fallback plan Do not delegate urgent blocking work that the parent agent should execute immediately unless explicitly requested by the parent agent.
© VoltAgent · MIT · VoltAgent/awesome-codex-subagents
Similar subagents
All subagents for Agent setup: memory, context, orchestration →multi-agent-coordinator by other authors comes first.