name	workflow-coding
description	Implement the smallest Smelter source change needed to turn RED tests GREEN under workflow gates.
version	0.55
type	workflow
consumes	.test. (RED) OR active_feedback
produces	src files changed
default_pattern	C
default_agent	executor
supports_patterns	["A","C"]
team_template	{"C":{"parallel_split_by":"module","agents":["executor"],"aggregator":"architect","conflict_resolver":"conflict-resolver","sync_point":"api-contract"}}
watchdog	E
can_delegate_to	["ui-ux-pro-max","copywriting","vercel-react-best-practices","designer"]
gate	{"pre":[{"tdd_cycle":"events[].type=test_red or legacy failing test_cycles[] exists"},{"allowed_in_mode":true}],"post":[{"src_changed":true},{"typecheck_clean":true},{"scoped_tests_pass":true}]}

workflow-coding

Implement the smallest source change that turns RED tests GREEN. This is not a terminal stage.

Announce: I'm using workflow-coding to turn RED tests GREEN. Next skill on pass: workflow-agent-review.

Required Inputs

RED evidence from workflow-write-test: real failing command exit, not ; echo "exit=$?" masking.
Or unresolved active_feedback targeting workflow-coding.
Current tasks.md or implementation-plan artifact relevant to the change.

Execution Checklist

Read the failing test and related source before editing.
Change only source files needed for the RED case or active feedback.
Do not edit tests in this stage. If tests are wrong or missing, route back to workflow-write-test.
Keep the patch minimal: no hardcoded secrets, no silent error swallowing, validate external input at boundaries, avoid deep nesting.
Run scoped tests for changed surface.
Run typecheck (tsc --noEmit or repo equivalent) when applicable.
If typecheck/scoped tests fail, fix and rerun; this is not a retry because code changed.

Gates

Pre-gate: RED evidence exists; otherwise route to workflow-write-test.
Post-gate: source changed, typecheck clean, scoped tests pass.
Full regression is opt-in at workflow-human-check; do not speculate it would pass.

Forbidden

Claiming done/fixed/complete before workflow-human-check returns complete.
Skipping workflow-agent-review because the change is small.
Skipping workflow-e2e for UI/CLI/API/hook/script behavior changes.
Asking the user whether to continue; the chain is autonomous until workflow-human-check.
Using TODOs or broad fallbacks to hide the defect.

Subagents

For independent /implement modules, dispatch executor subagents only when parallelization is safe. Give exact task text, file map, acceptance criteria, and excerpts from implementation-plan.md; require self-review and test output. Subagent self-review never replaces Smelter review/E2E/human-check.

Fail Routing

Missing RED evidence -> workflow-write-test
Scope mismatch in /implement -> workflow-implementation-plan
Scope mismatch in /fix -> workflow-investigate or mode upgrade
Persistent verification failure after code changes -> stay in workflow-coding with evidence

Terminal State

After post-gate passes, invoke workflow-agent-review. Do not stop, summarize as complete, or jump directly to E2E.

Más de este repositorio

mismo repositorio

smelter

Yusang-park/Smelter

Use when: User wants to run a SUBSTANTIAL multi-step coding task as a workflow (fix/implement a whole issue or feature, comprehensive review), CREATE workflows or commands, set up workflow automation, or manage workflow configuration. Triggers (run): "use workflow to", "run workflow", "with workflow", "workflow to", "workflow run", "use a workflow to", "run a workflow", and generic task-completion intents that imply a full coding deliverable: "실행해줘", "돌려줘", "구현해줘", "고쳐줘", "이슈 고쳐줘", "PR 만들어줘", "implement this", "fix this issue", "build this feature", "ship this". Triggers (create): "create a workflow", "write a workflow", "make a command", "author a workflow", "new workflow", "new command", "workflow yaml". Triggers (setup): "set up workflows", "install workflow automation", "how to use workflows", "configure workflows", "workflow setup", "get started with workflows". Triggers (config): "change my workflow config", "modify workflow config", "workflow

2026-05-262

agent-browser

Yusang-park/Smelter

Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.

2026-05-252

deep-interview

Yusang-park/Smelter

Smelter deep interview for planning-first requirement discovery before /brainstorm, /implement, or /fix execution

2026-05-252

release

Yusang-park/Smelter

Create a release from dev branch. Generates changelog entries from commits, bumps version, and creates a PR to main. TRIGGERS - Use this skill when user says: - "/release" - create a patch release (default) - "/release minor" - create a minor release - "/release major" - create a major release - "make a release", "cut a release", "ship it", "release to main"

2026-05-252

replicate-issue

Yusang-park/Smelter

Replicate and validate a GitHub issue by spinning up Smelter, analyzing the issue, and systematically testing all described symptoms using browser automation. Use when: User wants to reproduce a bug, validate a GitHub issue, confirm a reported problem, or investigate whether an issue is real before working on a fix. Triggers: "replicate issue", "reproduce issue", "validate issue", "confirm bug", "test issue", "can you reproduce", "try to replicate", "verify the bug". Capability: Checks out main, pulls latest, starts Smelter, reads the GitHub issue, then uses agent-browser to systematically test every symptom and produce a findings report. NOT for: Fixing issues (use /smelter or /exp-piv-loop:fix-issue), general UI testing (use /validate-ui).

2026-05-252

rulecheck

Yusang-park/Smelter

Autonomous rule adherence checker. Scans the codebase for rule violations, fixes the highest-impact ones in an isolated worktree, runs full validation, creates a PR. Uses memory to track progress across runs.

2026-05-252

workflow-coding

Implement the smallest source change that turns RED tests GREEN. This is not a terminal stage.

Announce: I'm using workflow-coding to turn RED tests GREEN. Next skill on pass: workflow-agent-review.

Required Inputs

RED evidence from workflow-write-test: real failing command exit, not ; echo "exit=$?" masking.

Or unresolved active_feedback targeting workflow-coding.

Current tasks.md or implementation-plan artifact relevant to the change.

Execution Checklist

Read the failing test and related source before editing.

Change only source files needed for the RED case or active feedback.

Do not edit tests in this stage. If tests are wrong or missing, route back to workflow-write-test.

Keep the patch minimal: no hardcoded secrets, no silent error swallowing, validate external input at boundaries, avoid deep nesting.

Run scoped tests for changed surface.

Run typecheck (tsc --noEmit or repo equivalent) when applicable.

If typecheck/scoped tests fail, fix and rerun; this is not a retry because code changed.

Gates

Pre-gate: RED evidence exists; otherwise route to workflow-write-test.

Post-gate: source changed, typecheck clean, scoped tests pass.

Full regression is opt-in at workflow-human-check; do not speculate it would pass.

Forbidden

Claiming done/fixed/complete before workflow-human-check returns complete.

Skipping workflow-agent-review because the change is small.

Skipping workflow-e2e for UI/CLI/API/hook/script behavior changes.

Asking the user whether to continue; the chain is autonomous until workflow-human-check.

Using TODOs or broad fallbacks to hide the defect.

Subagents

Fail Routing

Missing RED evidence -> workflow-write-test

Scope mismatch in /implement -> workflow-implementation-plan

Scope mismatch in /fix -> workflow-investigate or mode upgrade

Persistent verification failure after code changes -> stay in workflow-coding with evidence

Terminal State

After post-gate passes, invoke workflow-agent-review. Do not stop, summarize as complete, or jump directly to E2E.