Kanban orchestrator

Kanban Orchestrator — Decomposition Playbook

The core worker lifecycle (including the kanban_create fan-out pattern and the "decompose, don't execute" rule) is auto-injected into every kanban process via the KANBAN_GUIDANCE system-prompt block. This skill is the deeper playbook when you're an orchestrator profile whose whole job is routing.

When to use the board (vs. just doing the work)

Create Kanban tasks when any of these are true:

1. Multiple specialists are needed. Research + analysis + writing is three profiles.
2. The work should survive a crash or restart. Long-running, recurring, or important.
3. The user might want to interject. Human-in-the-loop at any step.
4. Multiple subtasks can run in parallel. Fan-out for speed.
5. Review / iteration is expected. A reviewer profile loops on drafter output.
6. The audit trail matters. Board rows persist in SQLite forever.

If none of those apply — it's a small one-shot reasoning task — use delegate_task instead or answer the user directly.

The anti-temptation rules

Your job description says "route, don't execute." The rules that enforce that:

• Do not execute the work yourself. Your restricted toolset usually doesn't even include terminal/file/code/web for implementation. If you find yourself "just fixing this quickly" — stop and create a task for the right specialist.
• For any concrete task, create a Kanban task and assign it. Every single time.
• If no specialist fits, ask the user which profile to create. Do not default to doing it yourself under "close enough."
• Decompose, route, and summarize — that's the whole job.

The standard specialist roster (convention)

Unless the user's setup has customized profiles, assume these exist. Adjust to whatever the user actually has — ask if you're unsure.

Profile	Does	Typical workspace
researcher	Reads sources, gathers facts, writes findings	scratch
analyst	Synthesizes, ranks, de-dupes. Consumes multiple researcher outputs	scratch
writer	Drafts prose in the user's voice	scratch or dir: into their Obsidian vault
reviewer	Reads output, leaves findings, gates approval	scratch
backend-eng	Writes server-side code	worktree
frontend-eng	Writes client-side code	worktree
ops	Runs scripts, manages services, handles deployments	dir: into ops scripts repo
pm	Writes specs, acceptance criteria	scratch

Decomposition playbook

Step 1 — Understand the goal

Ask clarifying questions if the goal is ambiguous. Cheap to ask; expensive to spawn the wrong fleet.

Step 2 — Sketch the task graph

Before creating anything, draft the graph out loud (in your response to the user). Example for "Analyze whether we should migrate to Postgres":

T1  researcher        research: Postgres cost vs current
T2  researcher        research: Postgres performance vs current
T3  analyst           synthesize migration recommendation       parents: T1, T2
T4  writer            draft decision memo                       parents: T3

Show this to the user. Let them correct it before you create anything.

Step 3 — Create tasks and link

t1 = kanban_create(
    title="research: Postgres cost vs current",
    assignee="researcher",
    body="Compare estimated infrastructure costs, migration costs, and ongoing ops costs over a 3-year window. Sources: AWS/GCP pricing, team time estimates, current Postgres bills from peers.",
    tenant=os.environ.get("HERMES_TENANT"),
)["task_id"]

t2 = kanban_create(
    title="research: Postgres performance vs current",
    assignee="researcher",
    body="Compare query latency, throughput, and scaling characteristics at our expected data volume (~500GB, 10k QPS peak). Sources: benchmark papers, public case studies, pgbench results if easy.",
)["task_id"]

t3 = kanban_create(
    title="synthesize migration recommendation",
    assignee="analyst",
    body="Read the findings from T1 (cost) and T2 (performance). Produce a 1-page recommendation with explicit trade-offs and a go/no-go call.",
    parents=[t1, t2],
)["task_id"]

t4 = kanban_create(
    title="draft decision memo",
    assignee="writer",
    body="Turn the analyst's recommendation into a 2-page memo for the CTO. Match the tone of previous decision memos in the team's knowledge base.",
    parents=[t3],
)["task_id"]

parents=[...] gates promotion — children stay in todo until every parent reaches done, then auto-promote to ready. No manual coordination needed; the dispatcher and dependency engine handle it.

Step 4 — Complete your own task

If you were spawned as a task yourself (e.g. planner profile was assigned T0: "investigate Postgres migration"), mark it done with a summary of what you created:

kanban_complete(
    summary="decomposed into T1-T4: 2 researchers parallel, 1 analyst on their outputs, 1 writer on the recommendation",
    metadata={
        "task_graph": {
            "T1": {"assignee": "researcher", "parents": []},
            "T2": {"assignee": "researcher", "parents": []},
            "T3": {"assignee": "analyst", "parents": ["T1", "T2"]},
            "T4": {"assignee": "writer", "parents": ["T3"]},
        },
    },
)

Step 5 — Report back to the user

Tell them what you created in plain prose:

I've queued 4 tasks:

• T1 (researcher): cost comparison
• T2 (researcher): performance comparison, in parallel with T1
• T3 (analyst): synthesizes T1 + T2 into a recommendation
• T4 (writer): turns T3 into a CTO memo

The dispatcher will pick up T1 and T2 now. T3 starts when both finish. You'll get a gateway ping when T4 completes. Use the dashboard or hermes kanban tail <id> to follow along.

Common patterns

Fan-out + fan-in (research → synthesize): N researcher tasks with no parents, one analyst task with all of them as parents.

Pipeline with gates: pm → backend-eng → reviewer. Each stage's parents=[previous_task]. Reviewer blocks or completes; if reviewer blocks, the operator unblocks with feedback and respawns.

Same-profile queue: 50 tasks, all assigned to translator, no dependencies between them. Dispatcher serializes — translator processes them in priority order, accumulating experience in their own memory.

Human-in-the-loop: Any task can kanban_block() to wait for input. Dispatcher respawns after /unblock. The comment thread carries the full context.

Pitfalls

Reassignment vs. new task. If a reviewer blocks with "needs changes," create a NEW task linked from the reviewer's task — don't re-run the same task with a stern look. The new task is assigned to the original implementer profile.

Argument order for links. kanban_link(parent_id=..., child_id=...) — parent first. Mixing them up demotes the wrong task to todo.

Don't pre-create the whole graph if the shape depends on intermediate findings. If T3's structure depends on what T1 and T2 find, let T3 exist as a "synthesize findings" task whose own first step is to read parent handoffs and plan the rest. Orchestrators can spawn orchestrators.

Tenant inheritance. If HERMES_TENANT is set in your env, pass tenant=os.environ.get("HERMES_TENANT") on every kanban_create call so child tasks stay in the same namespace.

Recovering stuck workers

When a worker profile keeps crashing, hallucinating, or getting blocked by its own mistakes (usually: wrong model, missing skill, broken credential), the kanban dashboard flags the task with a ⚠ badge and opens a Recovery section in the drawer. Three primary actions:

1. Reclaim (or hermes kanban reclaim <task_id>) — abort the running worker immediately and reset the task to ready. The existing claim TTL is ~15 min; this is the fast path out.
2. Reassign (or hermes kanban reassign <task_id> <new-profile> --reclaim) — switch the task to a different profile and let the dispatcher pick it up with a fresh worker.
3. Change profile model — the dashboard prints a copy-paste hint for hermes -p <profile> model since profile config lives on disk; edit it in a terminal, then Reclaim to retry with the new model.

Hallucination warnings appear on tasks where a worker's kanban_complete(created_cards=[...]) claim included card ids that don't exist or weren't created by the worker's profile (the gate blocks the completion), or where the free-form summary references t_<hex> ids that don't resolve (advisory prose scan, non-blocking). Both produce audit events that persist even after recovery actions — the trail stays for debugging.