/agent-task-splitter

name: agent-task-splitter
description: Use when about to run two or more subagents, workflow lanes, or delegated CLI executors on one goal. Gives each lane a role, a pinned model tier, a disjoint file scope, a bounded brief, and a capped return, and fixes the merge and verification step before launch. Not for research-domain routing that owns .research, .paper, Zotero, Obsidian, or NotebookLM workflows; use research-hub-multi-ai for those.

agent-task-splitter (fan-out planner)

Plan a fan-out before launching it. This skill plans; the host launches the lanes. The name stays `agent-task-splitter` so existing prompts and plans keep working.

Pick the mode

| Mode | Use when | Artifacts | |---|---|---| | Native (default) | All lanes run inside one host session: parallel or background subagents, workflow lanes, one-shot CLI delegates, with the user present | Lane table in the conversation and one brief per lane. No `.coord/` plan, packets, or checkpoint. | | Harness | A round spans sessions or tools, an unattended loop runs, work must resume after an interruption, a signed human gate is needed, or `AGENT_COLLAB_POLICY` is configured | `.coord/plan.yml` (schema v2) and `.ai/task_<NNN>_<slug>.md` packets; see [references/harness-plan.md](references/harness-plan.md) |

A configured policy always selects harness mode, and an unreadable configured policy fails closed. Native mode follows the host's own limits and permissions and adds no numeric limits of its own. Host mechanics are in [native lanes](../../docs/native-lanes.md).

Fan out only when

• two or more pieces are independent (none needs another's result) and each

has a checkable result; or

• implementation and independent review must be separate lanes.

Do not split exploratory debugging before the cause is known, a small coherent change, or work where every lane needs the same context in tight sequence. Doing it directly is the default.

The lane table

Write it before launching anything:

| Lane | Role | Executor | Model tier | May write | Must not touch | Acceptance | Return | Depends on | |---|---|---|---|---|---|---|---|---| | L1 | delegated-executor | subagent in a worktree | cheap | `src/a/**` | everything else | `pytest tests/a -q` | summary + paths | none | | L2 | delegated-executor | CLI delegate | cheap | `docs/x.md` | `src/**` | link check passes | summary + paths | none | | L3 | reviewer | read-only subagent | strong | nothing | `**/*` | verdict with evidence | verdict + findings | L1, L2 |

Rules:

1. Roles are `primary-agent`, `delegated-executor`, `reviewer`, and `synthesizer`, defined in the [public contract](../../docs/public-harness-contract.md). The executor and model columns are host-adapter choices, not public plan fields. 2. The orchestrator assigns the model tier; a lane never picks its own. Mechanical, pattern-following work goes to a cheap tier. Review, verdicts, root-cause analysis, security, governance, and anything ambiguous go to a strong tier. 3. Write scopes are disjoint. Two parallel writers never own the same file. If scopes might touch, isolate the writers or run them in sequence. 4. Every lane has at least one runnable or objectively checkable acceptance criterion. 5. Every lane has a capped return: a short summary plus file paths, never raw logs or full diffs. `agent-context-budget` covers sizing. 6. A lane cannot approve its own output. Author and reviewer are different lanes.

[references/task_splitter_heuristics.md](references/task_splitter_heuristics.md) helps when a task could fit more than one role or DAG shape.

Fix the merge step before launch

Write down, before any lane starts:

• who merges the results (normally the primary-agent) and in what order;
• the verification command run on the merged result, not only per lane;
• which `agent-acceptance-gate` presets the merged diff will trigger;
• commit boundaries: one reviewed commit per writing lane, or one commit

after the merged result passes. Read-only lanes need no commit.

Brief per lane

Lane <id>: <one bounded deliverable>

Repo/worktree: <absolute path> Role: <role>. Model tier: <tier>. Depends on: <lane ids and artifact paths>

Before editing, list the exact paths you may write and must not touch. Stop and report if they conflict with this brief.

May read: <paths> May write: <paths> Must not touch: <paths> External actions: none unless named here. Edits required: yes | unchanged output allowed because <reason>

Goal this lane serves: <the outcome of the whole job, one line> Already decided (do not reopen): <decisions, copied word for word>

Acceptance: <runnable check> Return (capped): status; short summary; files changed; the acceptance command and the last lines of its actual output; other checks run and results; risks; blockers. Put long output in a file and return its path.

If the job already has a plan header (for example from the `aim-higher` skill), copy its outcome, outside check, not-doing and already-decided lines into these fields instead of rewriting them. Without one, fill the fields yourself.

Drift guards

• Copy identifiers (slugs, paths, field names, version strings) verbatim into

the brief. Do not paraphrase them.

• Record the pre-lane baseline (commit or file hashes) so the result can be

attributed. File times and a process exit status do not prove the work was done; use the [task-content evidence review](../../docs/task-content-evidence.md).

• State whether edits are required. A lane that reports success without

changing its required files has failed unless the brief allowed unchanged output.

• Check each return against its brief when it arrives, not only at the end: a

"done" without the acceptance command's actual output stays unverified, and a result that departs from the brief's steps or decided items is drift even when its own checks pass.

• Missing, null, failed,