Architect
Design before implementing. Sketch types, function signatures, class shapes, and module boundaries with not implemented bodies and pseudocode. Synthesize across multiple independent perspectives, then fill in code against the chosen sketch. If implementation proves the sketch wrong, throw it out and redesign.
Start
Open a todolist with one entry per phase before starting. Autonomous mode without checkpoints needs the list to show phase position and keep phases from silently disappearing.
- Ground
- Sketch
- Agree
- Implement
- Scrap
Phase A: Ground the problem
Build a real mental model of every system the new code touches. Run the how skill over the relevant subsystems. Critique mode if existing structure is the constraint or the design must push back on it.
Naming a file isn't grounding. Produce the traced model how prescribes. If the design redefines ownership or layering, also run the why skill on the existing shape so the rationale becomes a constraint, not a guess.
Skip Phase A only when the work is genuinely greenfield with no surrounding system to integrate.
Phase B: Sketch
Fan out. Spawn three or four independent subagents in one message, each with references/runner-prompt.md as its prompt plus the design-sketch task and the Phase A grounding artifacts. Each candidate produces a design package shaped per references/rationale-template.md: the caller's usage written first, then the type sketch, function signatures, module map, and prose rationale derived from it.
Isolation is what makes the fan-out worth running. Give each runner its own working directory — a git worktree when available, otherwise a per-runner subdirectory under the sketch dir — and do not let one candidate see another's output before all have returned. Candidates that can see each other converge, and convergence is the failure mode.
Diversity is the other half. If the harness lets you vary the model, reasoning effort, or thinking budget per subagent, do — divergent candidates are the whole point. If it does not, vary the framing instead: give each runner a different explicit bias (optimise for the simplest public surface / for the fewest moving parts / for the most direct data flow / for the shape that survives the next requirement). One model asked four different questions still beats one model asked the same question four times.
Design it twice. Require at least two structurally distinct candidates before synthesis, even when the first looks sufficient. This is the exhaust-the-design-space principle made concrete. Whole-shape alternatives, not point fixes inside one shape.
Screen every candidate against references/design-red-flags.md before synthesis. Reject or revise shallow modules, information leakage, temporal decomposition, and pass-through methods.
Compare viable candidates on interface depth. Prefer the design that hides more complexity behind a smaller, simpler public surface. A rich interface can keep call chains short by concentrating capability instead of scattering it across layers.
Synthesize the candidates yourself into one design package: pick a base, graft what the others did better, and reject the rest explicitly. Record which candidate became the base and why in the rationale's "Synthesis decision" section. A synthesis that takes everything from one candidate is a legitimate outcome — say so rather than manufacturing a blend.
Phase C: Agree (opt-in)
Default: proceed directly to implementation with the synthesized design. No human checkpoint.
Opt in to a checkpoint when the invoker explicitly asks: "/architect with checkpoint," "stop and show me before implementing," or similar. Then surface the synthesized design and pause for sign-off.
The synthesis can ship as its own commit either way. That's the "scaffold first" mode of the foundational-thinking principle; subsequent commits read as filling in bodies against a stable contract. Planned and scoped breakage during fill-in is fine, per the outcome-oriented-execution principle. For adversarial pressure on the design before implementing, spawn one read-only subagent whose brief is to attack the synthesized sketch — find the requirement it cannot absorb, the invariant it leaves unencoded, the caller it makes awkward — and answer its findings before Phase D.
If the human pushes back on the shape (in a checkpoint or after the fact), treat that as Phase A evidence. Re-ground and re-run Phase B before writing more code.
Phase D: Implement against the sketch
Replace not implemented bodies with code, pseudocode with logic. The synthesized sketch is the contract.
Deviations from the sketch are signal worth surfacing, not friction to absorb silently. If a function needs a parameter the sketch didn't anticipate, ask whether the sketch was wrong, the requirement was missed, or the implementation is overreaching. Surface it; don't bolt it on.
Phase E: Scrap when the architecture is wrong
If implementation keeps producing friction the sketch can't absorb, throw the sketch out. Don't bolt fixes onto a wrong design, per the redesign-from-first-principles and fix-root-causes principles.
The signal is a pattern, not single instances. Tells:
- The same shape of workaround appearing repeatedly across unrelated code.
- Multiple unrelated edge cases that all need special-case branches.
- Types that need escape hatches (
any, casts, optional fields always set in practice) to compile. - The "we need a lock" reflex when the sketch said the state wasn't shared.
- Callers having to know the abstraction's internal rules to use it.
- Two or more independent Phase D deviations of the same shape across the implementation. Surfacing deviations is Phase D's job; a repeated pattern of them is Phase E's trigger.
Use judgment. A few edge cases don't condemn an architecture. Some problems are legitimately complex; complexity in the data is not complexity in the design. The rewrite signal is repeated friction of the same shape, not single hard cases.
When you scrap:
- Re-run the how skill over what's been built. The implementation lessons enter the new design as inputs, not vibes.
- Redesign as if the new constraints had been day-one assumptions, per redesign-from-first-principles.
- Subtract before adding, per the subtract-before-you-add principle. The new sketch should be smaller than the old one before it grows.
- Return to Phase B and re-run the fan-out.
Outputs
The caller's usage is written first and the type sketch derived from it. One file with new types and signatures for small changes; module map plus type definitions for larger work. The rationale ships alongside, shaped per references/rationale-template.md, including the usage sketch and the synthesis decision.