atelier-orchestrator

---
name: atelier-orchestrator
description: >
  Skill routing and workflow orchestration. Selects Inline Plan or Spec-backed Plan, routes
   to the correct workflow skill, and manages transitions between phases. Use when starting
  any conversation or task to determine which planning mode and skill apply.
user-invocable: false
---

<EXTREMELY-IMPORTANT>
If you think there is even a 1% chance a skill might apply to what you are doing, you ABSOLUTELY
MUST invoke the skill.

IF A SKILL APPLIES TO YOUR TASK, YOU DO NOT HAVE A CHOICE. YOU MUST USE IT.

This is not negotiable. This is not optional. You cannot rationalize your way out of this.
</EXTREMELY-IMPORTANT>

# Planning Workflow

You have skills. You MUST use them. Not "should." Not "when convenient." MUST.

Invoke relevant skills BEFORE any response or action. Even a 1% chance a skill might apply
means you invoke it. If an invoked skill turns out to be wrong for the situation, you don't
need to follow it. But you must check.

## The One Rule

**When an implementation choice exists, never write code until the human has reviewed and
approved a written plan.** Exact mechanical changes may use the exception below.

The plan may be inline in the conversation or backed by persisted spec artifacts. Select
the mode before invoking an artifact-producing skill, state the choice with one brief
reason, and let the human override it.

## Governing Principles

Invoking a skill selects relevant guidance. It does not make every pattern in every loaded
skill part of the solution. Existing source, explicit requirements, and the smallest correct
change determine which patterns apply.

Begin with the current design and the minimum behavior-preserving change. Product,
infrastructure, and architectural expansion require explicit scope. Keep **ponytail** active
throughout development when it is available or required by the repository. Keep
**writing-clearly-and-concisely** active whenever writing or editing prose for humans. Before
returning or persisting that prose, apply **humanizer** in embedded mode.

## Planning Modes

### Inline Plan (default)

Use for bounded, well-understood work, including ordinary features, bug fixes, refactors,
configuration changes, and multi-file changes. Route to **spec-plan** with Inline mode
explicitly selected. It presents a concise plan in conversation, creates no planning artifacts
or tracker entries, and stops for approval. Approval ends the `spec-plan` invocation. A later,
explicit implementation request may implement the plan directly; do not invoke
`spec-implement`, `spec-finish`, or `code-subagents` for Inline work.

Direct Inline implementation must:

- Avoid main/master unless the human explicitly permits it
- Stay within the approved scope and files
- Run every validation item in the approved plan
- Review the final diff for unintended changes
- Do not commit, push, or open a PR without explicit human instruction

When in doubt, choose Inline Plan. File count and estimated duration alone do not make work
spec-worthy.

### Spec-backed Plan

Use when at least one concrete signal exists:

- The human explicitly requests a spec or durable planning artifacts
- Requirements or boundaries need discovery before planning
- The work materially changes architecture, public contracts, data models, or cross-system behavior
- Multiple dependent subsystems need decomposition and dependency tracking
- The plan must survive the conversation for handoff, audit, or multi-person coordination

Route through **spec-brainstorm**. This mode produces:

```
design.md    ← spec-brainstorm (requirements + research + architecture)
plan.json    ← spec-plan (tasks, dependencies)
```

Each skill produces only its declared output and stops. The human starts the next phase with a
separate request.

Do not infer substantial work from size alone. If no concrete signal applies, use Inline Plan.

## Skill Routing

```
atelier-orchestrator → Select and announce planning mode
spec-brainstorm   → Spec-backed discovery + design → only design.md → stop
spec-plan         → Inline conversation or only plan.json → stop
spec-implement    → Execute approved Spec-backed Plans; track tasks
spec-finish       → Post-implementation validation for Spec-backed Plans
code-subagents    → Parallel dispatch for Spec-backed Plan tasks
```

### Ordinary bounded work

```
spec-plan (Inline mode) → approval → STOP
new implementation request → implement directly
```

### Substantial work or explicit spec request

```
spec-brainstorm → design.md → STOP
new spec-plan request → plan.json → STOP
new spec-implement request → implementation → spec-finish
```

### Truly mechanical change

A typo or exact single-line replacement may skip planning when there is no implementation
choice to review. If any choice exists, use an Inline Plan.

## Hard Transitions

| After completing... | Next step |
|---------------------|-----------|
| spec-brainstorm | Stop; wait for a separate spec-plan request |
| approved Inline Plan | Stop; wait for a separate implementation request |
| approved Spec-backed Plan | Stop; wait for a separate spec-implement request |
| spec-implement | spec-finish |

Do not begin implementation without an approved written plan. Spec-backed work must not
jump from research to implementation.

`spec-brainstorm` and `spec-plan` are terminal within their invocation. They must not chain into
the next workflow skill, even when the original request asks for multiple phases at once.

## Iteration Patterns

The workflow is not purely linear. Expect backflows:

### Research → Research (discovery loop)
- Brainstorm reveals new requirements → loop back to discovery phase
- Human adds new scope mid-brainstorm → continue discovery

### Plan → Research (design flaw)
- Planning reveals design assumptions are wrong → back to brainstorm
- Tasks can't be decomposed without more context → back to brainstorm

### Implement → Plan (material changes)
- A minor deviation may be recorded inline and reflected in the plan when it does not change
  approved behavior, scope, architecture, public contracts, or major dependencies.
- Material missing work or an unplanned dependency → revise the active plan and get approval.

### Implement → Research (fundamental issue)
- Inline planning or implementation develops substantial needs → ask the human whether to
  switch to a Spec-backed Plan
- Implementation reveals spec-backed design is fundamentally wrong → back to brainstorm

### Finish → Implement (bugs found)
- Validation finds bugs → back to implement
- Tests failing → back to implement

### When to escalate
If you loop 2+ times on the same issue, stop and ask the human:

> "We've looped on [issue] twice. Should we reconsider the approach?"

## Red Flags — You Are Rationalizing

| What you're thinking | Why it's wrong |
|----------------------|----------------|
| "This is too simple for a persisted spec" | Use an Inline Plan instead |
| "I already know how to do this" | Knowing how ≠ having the human's approval for how |
| "The human seems impatient" | Wasting time on wrong code is worse than planning |
| "I'll just do a quick prototype" | Prototypes become production. Plan it. |
| "I need to explore the code first" | Gather enough context for an Inline Plan; use brainstorm only when discovery is substantial |
| "Let me just fix this one thing" | One thing becomes three. Plan it. |
| "I can plan in my head" | Plans in your head can't be reviewed or annotated |
| "This is just a refactor" | Refactors still need at least an Inline Plan |
| "I'll write the plan after" | Post-hoc plans are fiction. Plan before. |
| "I need more context first" | Skills tell you HOW to gather context. Check first. |
| "The spec workflow is overkill" | It may be. Default to an Inline Plan unless substantial-work signals exist. |
| "I know what that skill says" | Skills evolve. Read current version. Invoke it. |
| "This feels productive" | Undisciplined action wastes time. Skills prevent this. |

## Skill Types

**Process skills** (spec-brainstorm, spec-plan, spec-implement, spec-finish): Follow exactly.
Don't adapt away discipline.

**Knowledge skills** (oracle-grill-me, oracle-domain-modelling): Adapt principles to
context. These inform decisions within the workflow.

**Discipline skills** (oracle-debug): Strict methodology that must be followed
exactly — no adaptation that bypasses root-cause investigation.

Process skills come first. Knowledge skills get invoked by process skills when needed.

## User Instructions

"Add X" or "Fix Y" doesn't mean skip workflows. Instructions say WHAT, not HOW.
The skills define HOW.