coding-flow

<coding_flow>

<description_and_purpose>

Problem: Unstructured coding leads to scope drift, missing validation, autonomous runaway, and misaligned deliverables. Solution: Sequential workflow with reviewer gates, HITL gates, subagent delegation, and skill-driven execution scaled per Request size classification. Validation: Each phase produces verifiable outputs; reviewer catches issues before user; HITL gates prevent autonomous runaway; final validation confirms implementation matches approved intent. Lightweight variant: a single architect pass produces discovery, design, specs, and plan; one reviewer gate and one HITL gate approve all of it before implementation.

</description_and_purpose>

<workflow_phases>

1. All Rosetta prep steps MUST be FULLY completed
2. MUST USE SKILL load-project-context (required: all), orchestration (all except trivial), hitl (all, unless No HITL or Fully Autonomous)
3. MUST ALWAYS use todo tasks ledger, ASAP. Phases are sequential. Independent tasks can run in parallel.
4. No rush, take your time, MUST FOLLOW WORKFLOW ENTIRELY, no skipping, if in doubt - select the safest / longest path, no deviation from the workflow is allowed
5. When debugging is needed, INVOKE SUBAGENT engineer with debugging skill to save LLM context
6. INVOKE SUBAGENT executor for building, running tests, installing packages, and similar mechanical actions.
7. MUST just-in-time load each phase's skills
8. If workflow is for REQUIREMENTS, MUST USE SKILL requirements-use and LOAD all affected requirements. Use refs to requirements for subagents.
9. If /goal is set repeat phases 4-7 postponing final_validation until goal is met.
10. If migrate/modernize: implementation phase MUST use tiny batches ONLY (1-3 files), never bulk-read (other phases may); specs/plan enforce; FS-copy RECOMMENDED; no behavior change/new code; mirror source; subagents same; REQUIRED TO log started/completed; Use impl subagents like MAP-REDUCE;
11. Run architect subagent with required model in the background and consult with it if already supported or prefer advisor if already available
12. Coding workflow state MUST be saved to agents/TEMP/<FEATURE>/coding-flow-state.md file.
13. If context already contains request for TDD => implement properly red and green adjusting this workflow, including by moving execution of phases tests and review_tests to be after user_review_design and before implementation.

<solution_design phase="1" applies="ALL" subagent="architect" role="Architect producing discovery, design, specs, and plan in one pass" subagent_required_model="GPT-5.6 Sol, Claude Opus 5, Gemini 3.7 Flash">

Execute strongly in the specified order. A step MUST NOT start before the previous step is complete.

1. Step 1: Discover project context, affected and related code, dependencies, constraints, and requirements; derive the architecture requirements that address the user request fully. Input: user request + CONTEXT.md + ARCHITECTURE.md + IMPLEMENTATION.md. Do not stop until 100% clear.
2. Step 2: Design 3 best architecture solutions on high level with pro/cons analysis.
3. Step 3: Define the best solution, but concise, phrase-terse, compressed, etc. Record Steps 1-3 as concise architecture-notes.md in FEATURE PLAN folder.
4. Step 4: Once done, USE SKILL tech-specs (load JIT) to create plans/<FEATURE>/<FEATURE>-SPECS.md. Specs own WHAT.
5. Step 5: Once done, USE SKILL planning (load JIT) to create plans/<FEATURE>/<FEATURE>-PLAN.md. Plan owns HOW. Target: 100% clarity. Cross-reference specs, never duplicate them.
6. Required skills: reasoning, tech-specs, planning
7. Recommended skills: questioning, codemap (structural project discovery)
8. If REQUIREMENTS in use: requirements-use skill is required. Plan/Specs must have pointers to requirements identifiers.
9. Additionally discover existing libraries, packages, and search web for similar problems/tasks (if this make sense)
10. SMALL: output specs and plan as message, no files. MEDIUM: concise. LARGE: full.
11. Update coding-flow-state.md