chhoumann/t3-pstack/swarm

Swarm

Start

Open a todolist with one entry per phase before launching anything.

1. Frame
2. Fan out
3. Aggregate
4. Report

Phase A: Frame

1. State the done predicate and the artifact or report the swarm must return.
2. Choose the shape. Partition into slices, race N workers on identical briefs, or mix both. For a race or mixed shape, declare first pass, rank all, or best-of before spawning.
3. Pick the worker model from the swarm workers line in the model sheet. If the sheet or that line is missing, use the default in Models. For auto or inherit-parent, omit target so the workers run on the parent model. If delegate_task rejects an entry, use the default and say so. If it rejects the default, use another target of the same family from orchestrator_capabilities. For a model race, name each arm's model up front. Pick each subagent's reasoning level and fast or standard speed for the job, per the Subagents section of the poteto-mode skill.
4. Give each worker its own writable output when it writes. When workers verify or measure commits, each brief names the exact SHAs. A measurement brief also names the method (sample count, what one sample is, order). The worker records both in its result.

Phase B: Fan out

Spawn all N workers in one response as delegate_task calls with mode: "async" and the step 3 target, left unset for auto or inherit-parent. Every worker runs on this machine, so a worker that writes gets its own git worktree, named by absolute path in its brief as the only place it writes.

When a worker must start from a non-default branch, create its worktree from that branch.

Every brief stands alone. Include the goal, scope, exact slice or race arm, how to verify, and what to report. Reports use PASS, ISSUES, or BLOCKED with evidence. A worker that can prove a defect reports ISSUES and lists every issue it can prove, not only the first.

If a worker drops out, proceed with N-1 and note it.

Phase C: Aggregate

Read the terminal results. Drop a result that does not record the SHAs and method its brief names, and respawn that worker once. After a second miss, record a gap. A gap does not count as a pass. For coverage, every required slice needs a result. For a race, apply the selection rule declared up front. Use first pass, rank all, or best-of. Do not paste raw worker dumps.

Phase D: Report

Return one consolidated in-chat report with a result table, issue one-liners, gaps or dropouts, and the race rule when used.

Models

Role defaults on T3, stamped from plugins/pstack/models.json. A role line in the model sheet overrides each one. The sheet is ~/.config/t3-pstack/models.md ($XDG_CONFIG_HOME/t3-pstack/models.md when that is set), one per machine, and /setup-pstack writes it. When it is missing, read t3-pstack-models.md in the host's config directory ($CLAUDE_CONFIG_DIR or ~/.claude on Claude Code, $CODEX_HOME or ~/.codex on Codex). An entry is <providerInstanceId>/<modelId> plus <optionId>=<value> pairs, and it maps onto delegate_task's target, with the provider instance as providerInstanceId, the model as model, and the pairs as options. It splits at its first /, so a model ID with its own /, such as OpenCode's <provider>/<model>, is written whole after it. Option IDs differ per provider, so take them from orchestrator_capabilities. inherit-parent or auto omits target.

• swarm workers: claudeAgent/claude-opus-5-5 effort=xhigh