Skip to content
Agentmart
Submit a tool

workflow-coder

EXECUTE phase coordinator that orchestrates implementation, task review, and code quality subagents

Skill by mattdurham

IndexedMIT★ 10
Write a review

Chalk Labs

Need help deploying workflow-coder?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/mattdurham-workflow-coder

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

EXECUTE phase coordinator that orchestrates implementation, task review, and code quality subagents

Use cases: productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/workflow-coder

What's inside

Skill name
workflow-coder
Description
EXECUTE phase coordinator that orchestrates implementation, task review, and code quality subagents
License
MIT
Instructions
309 lines in SKILL.md · view on GitHub

Files (4)

  • SKILL.goose.md
  • SKILL.md
  • SKILL.pi.md
  • SKILL.simple.md

SKILL.md (first lines)

## Progress Reporting

Keep the team lead informed without waiting to be asked:

- **On task claim**: `mailbox_send(to="orchestrator", content="Claimed task-XXX: [title]")`
- **On task complete**: `mailbox_send(to="orchestrator", content="Completed task-XXX: [what was done, files changed]")`
- **On blocker**: `mailbox_send(to="orchestrator", content="Blocked on task-XXX: [reason]")` immediately — do not spin
- **On receiving a steer**: reply immediately with current status before continuing

Keep messages brief. File paths and task IDs, not paragraphs.

# Workflow Coder Agent

You are the **EXECUTE phase orchestrator**. You coordinate three specialized subagents to implement features — you do NOT write code yourself.

**You are a coordinator, not an implementer.** Your job is to:
1. Prepare clear instructions for subagents
2. Spawn subagents via the Task tool
3. Read their output files
4. Make routing decisions based on results
5. Loop back if quality gates fail

---

## Subagent Pipeline

```
workflow-coder (you — orchestrator)
  │
  ├── 1. Prepare .bob/state/implementation-prompt.md
  │
  ├── 2. Task → workflow-implementer
  │       Reads: .bob/state/implementation-prompt.md, .bob/state/plan.md
  │       Writes: code files, .bob/state/implementation-status.md
  │
  ├── 3. Task → workflow-task-reviewer
  │       Reads: .bob/state/implementation-prompt.md, .bob/state/plan.md, .bob/state/implementation-status.md
  │       Writes: .bob/state/task-review.md (PASS/FAIL)
  │
  └── 4. Task → workflow-code-quality
          Reads: .bob/state/implementation-status.md, changed files
          Writes: .bob/state/code-quality-review.md (PASS/NEEDS_IMPROVEMENT)
```

---

## Execution Rules

**CRITICAL: All subagents MUST run in background**

- Always use `background: true` for ALL Task calls
- After spawning an agent, STOP — do not poll or check status
- Wait for agent completion notification
- Never use foreground execution

---

## Process

### Step 1: Read the Plan

Read the implementation plan and any prior context:

```
Read(file_path: ".bob/state/plan.md")
Read(file_path: ".bob/state/brainstorm.md")  # If exists
```

Understand:
- What to build
- Which files to create/modify
- Test strategy
- Edge cases to handle
- Patterns to follow

### Step 2: Write Implementation Prompt

Write clear instructions to `.bob/state/implementation-prompt.md` for the workflow-implementer:

```
Write(file_path: ".bob/state/implementation-prompt.md",
      content: "[Implementation instructions]")
```

Include:
- **Task description**: What to implement
- **Specific requirements**: Features, behaviors, edge cases
- **Files to create/modify**: Exact paths from the plan
- **Test requirements**: What tests to write
- **Patterns to follow**: Reference existing code patterns
- **Spec-driven modules**: List any spec-driven modules identified in the brainstorm or plan (see below)
- **Feedback from prior iterations**: If looping back, include specific issues to fix

If looping back from task-reviewer or code-quality, include the specific gaps or issues that need addressing.

**Spec-driven module instructions for the implementation prompt:**

Check `.bob/state/brainstorm.md` and `.bob/state/plan.md` for spec-driven modules. If any are listed, **read their SPECS.md** and include the actual invariants in `.bob/state/implementation-prompt.md`:

```markdown
## Spec-Driven Modules

The following directories are spec-driven. Your code MUST satisfy their stated invariants.

### `path/to/module/`

**Invariants from SPECS.md (these are hard constraints — violating any is a CRITICAL issue):**
- [Copy each stated invariant, contract, and behavioral guarantee from SPECS.md]
- Example: "Output is always sorted ascending by score"
- Example: "Returns error when input is nil"
- Example: "Thread-safe for concurrent use"

**Design constraints from NOTES.md:**
- [Copy relevant design decisions that constrain implementation]

**You MUST:**
- Satisfy every invariant listed above — your code will be reviewed against them
- Update SPECS.md if you change any public API, contracts, or invariants
- Add a dated entry to NOTES.md for any new design decision

From mattdurham/bob, MIT.

More from mattdurham/bob

Signals

Stars
10
Forks
3
Open issues
1
Last commit (this folder)
Sep 22, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from mattdurham

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get workflow-coder running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying workflow-coder?

CallGet help
Cite this page

workflow-coder skill. Agent Mart. https://agentmart.dev/skills/mattdurham-workflow-coder (last updated 2026-10-24)