Skip to main content
skill CX Codex

subagent-driven-development · nousresearch — Claude Code skill: how to install

subagent-driven-development runs an implementation plan task by task. Each task goes to a fresh subagent via delegate_task, then gets a spec review and a code quality review.

MIT 106 installs ★ 265,216 updated 2026-10-25

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add NousResearch/hermes-agent --skill subagent-driven-development -a claude-code
Open on GitHub

What subagent-driven-development does

The skill reads the plan file once, pulls out every task, and builds a todo list. Each task then goes to its own fresh implementer subagent started with delegate_task. The full task text and project context are pasted into the prompt, so the subagent never opens the plan itself.

Two reviewers check each finished task, in order. The first compares the code to the task spec and answers PASS or lists the gaps. The second judges code quality and returns APPROVED or REQUEST_CHANGES. Quality review does not start until the spec review passes, and the next task waits until both are clean.

After the last task a final reviewer checks that the pieces fit together, then the full test suite runs and the work is committed. Tasks should be small, 2-5 minutes each. Two implementers must not work on tasks that touch the same files. If a subagent fails, a new fix subagent takes over instead of the main session.

What it helps with

  • work through a finished plan without piling context into one session
  • check code against the task spec before arguing about style
  • catch extra code nobody asked for
  • get an APPROVED or REQUEST_CHANGES verdict on every task
  • confirm before merge that the plan's parts fit together

Example prompts for the agent

  • Execute the plan in docs/plans/feature-plan.md with the subagent-driven-development skill
  • Give each task in the plan to a fresh subagent and run a spec review and a quality review after each one
  • Use subagent-driven-development: one subagent per task, then a final integration review

How to install subagent-driven-development

1Skills CLI (Vercel)

npx skills add NousResearch/hermes-agent --skill subagent-driven-development -a claude-code

2Manually: put the files in place

~/.claude/skills/subagent-driven-development/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/subagent-driven-development/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about subagent-driven-development

How do I install subagent-driven-development in Claude Code?

Run npx skills add NousResearch/hermes-agent --skill subagent-driven-development -a claude-code. The skill file ends up at ~/.claude/skills/subagent-driven-development/SKILL.md.

What does subagent-driven-development do?

It takes a ready plan and hands each task to a fresh subagent through delegate_task. Every task gets two reviews, and the whole implementation gets one last check at the end.

What do I need before running this skill?

An implementation plan, either from the plan skill or written from your requirements. The skill assumes the tasks are mostly independent of each other.

Source text

SKILL.md · MIT
---
name: subagent-driven-development
description: "Execute plans via delegate_task subagents (2-stage review)."
version: 1.1.0
author: Hermes Agent (adapted from obra/superpowers)
license: MIT
platforms: [linux, macos, windows]
metadata:
  hermes:
    tags: [delegation, subagent, implementation, workflow, parallel]
    related_skills: [requesting-code-review, test-driven-development]
---

# Subagent-Driven Development

## Overview

Execute implementation plans by dispatching fresh subagents per task with systematic two-stage review.

**Core principle:** Fresh subagent per task + two-stage review (spec then quality) = high quality, fast iteration.

## When to Use

Use this skill when:
- You have an implementation plan (from the `plan` skill or user requirements)
- Tasks are mostly independent
- Quality and spec compliance are important
- You want automated review between tasks

**vs. manual execution:**
- Fresh context per task (no confusion from accumulated state)
- Automated review process catches issues early
- Consistent quality checks across all tasks
- Subagents can ask questions before starting work

## The Process

### 1. Read and Parse Plan

Read the plan file. Extract ALL tasks with their full text and context upfront. Create a todo list:

```python
# Read the plan
read_file("docs/plans/feature-plan.md")

# Create todo list with all tasks
todo([
    {"id": "task-1", "content": "Create User model with email field", "status": "pending"},
    {"id": "task-2", "content": "Add password hashing utility", "status": "pending"},
    {"id": "task-3", "content": "Create login endpoint", "status": "pending"},
])
```

**Key:** Read the plan ONCE. Extract everything. Don't make subagents read the plan file — provide the full task text directly in context.

### 2. Per-Task Workflow

For EACH task in the plan:

#### Step 1: Dispatch Implementer Subagent

Use `delegate_task` with complete context:

```python
delegate_task(
    goal="Implement Task 1: Create User model with email and password_hash fields",
    context="""
    TASK FROM PLAN:
    - Create: src/models/user.py
    - Add User class with email (str) and password_hash (str) fields
    - Use bcrypt for password hashing
    - Include __repr__ for debugging

    FOLLOW TDD:
    1. Write failing test in tests/models/test_user.py
    2. Run: pytest tests/models/test_user.py -v (verify FAIL)
    3. Write minimal implementation
    4. Run: pytest tests/models/test_user.py -v (verify PASS)
    5. Run: pytest tests/ -q (verify no regressions)
    6. Commit: git add -A && git commit -m "feat: add User model with password hashing"

    PROJECT CONTEXT:
    - Python 3.11, Flask app in src/app.py
    - Existing models in src/models/
    - Tests use pytest, run from project root
    - bcrypt already in requirements.txt
    """,
    toolsets=['terminal', 'file']
)
```

#### Step 2: Dispatch Spec Compliance Reviewer

After the implementer completes, verify against the original spec:

```python
delegate_task(
    goal="Review if implementation matches the spec from the plan",
    context="""
    ORIGINAL TASK SPEC:
    - Create src/models/user.py with User class
    - Fields: email (str), password_hash (str)
    - Use bcrypt for password hashing
    - Include __repr__

    CHECK:
    - [ ] All requirements from spec implemented?
    - [ ] File paths match spec?
    - [ ] Function signatures match spec?
    - [ ] Behavior matches expected?
    - [ ] Nothing extra added (no scope creep)?

    OUTPUT: PASS or list of specific spec gaps to fix.
    """,
    toolsets=['file']
)
```

**If spec issues found:** Fix gaps, then re-run spec review. Continue only when spec-compliant.

#### Step 3: Dispatch Code Quality Reviewer

After spec compliance passes:

```python
delegate_task(
    goal="Review code quality for Task 1 implementation",
    context="""
    FILES TO REVIEW:
    - src/models/user.py
    - tests/models/test_user.py

    CHECK:
    - [ ] Follows project conventions and style?
    - [ ] Proper error handling?
    - [ ] Clear variable/function names?
    - [ ] Adequate test coverage?
    - [ ] No obvious bugs or missed edge cases?
    - [ ] No security issues?

    OUTPUT FORMAT:
    - Critical Issues: [must fix before proceeding]
    - Important Issues: [should fix]
    - Minor Issues: [optional]
    - Verdict: APPROVED or REQUEST_CHANGES
    """,
    toolsets=['file']
)
```

**If quality issues found:** Fix issues, re-review. Continue only when approved.

#### Step 4: Mark Complete

```python
todo([{"id": "task-1", "content": "Create User model with email field", "status": "completed"}], merge=True)
```

### 3. Final Review

After ALL tasks are complete, dispatch a final integration reviewer:

```python
delegate_task(
    goal="Review the entire implementation for consistency and integration issues",
    context="""
    All tasks from the plan are complete. Review the full implementation:
    - Do all components work together?
    - Any inconsistencies between tasks?
    - All tests passing?
    - Ready for merge?
    """,
    toolsets=['terminal', 'file']
)
```

### 4. Verify and Commit

```bash
# Run full test suite
pytest tests/ -q

# Review all changes
git diff --stat

# Final commit if needed
git add -A && git commit -m "feat: complete [feature name] implementation"
```

## Task Granularity

**Each task = 2-5 minutes of focused work.**

**Too big:**
- "Implement user authentication system"

**Right size:**
- "Create User model with email and password fields"
- "Add password hashing function"
- "Create login endpoint"
- "Add JWT token generation"
- "Create registration endpoint"

## Red Flags — Never Do These

- Start implementation without a plan
- Skip reviews (spec compliance OR code quality)
- Proceed with unfixed critical/important issues
- Dispatch multiple implementation subagents for tasks that touch the same files
- Make subagent read the plan file (provide full text in context instead)
- Skip scene-setting context (subagent needs to understand where the task fits)
- Ignore subagent questions (answer before letting them proceed)
- Accept "close enough" on spec compliance
- Skip review loops (reviewer found issues → implementer fixes → review again)
- Let implementer self-review replace actual review (both are needed)
- **Start code quality review before spec compliance is PASS** (wrong order)
- Move to next task while either review has open issues

## Handling Issues

### If Subagent Asks Questions

- Answer clearly and completely
- Provide additional context if needed
- Don't rush them into implementation

### If Reviewer Finds Issues

- Implementer subagent (or a new one) fixes them
- Reviewer reviews again
- Repeat until approved
- Don't skip the re-review

### If Subagent Fails a Task

- Dispatch a new fix subagent with specific instructions about what went wrong
- Don't try to fix manually in the controller session (context pollution)

## Efficiency Notes

**Why fresh subagent per task:**
- Prevents context pollution from accumulated state
- Each subagent gets clean, focused context
- No confusion from prior tasks' code or reasoning

**Why two-stage review:**
- Spec review catches under/over-building early
- Quality review ensures the implementation is well-built
- Catches issues before they compound across tasks

**Cost trade-off:**
- More subagent invocations (implementer + 2 reviewers per task)
- But catches issues early (cheaper than debugging compounded problems later)

## Integration with Other Skills

### With plan

This skill EXECUTES plans created by the `plan` skill:
1. User requirements → plan → implementation plan
2. Implementation plan → subagent-driven-development → working code

### With test-driven-development

Implementer subagents should follow TDD:
1. Write failing test first
2. Implement minimal code
3. Verify test passes
4. Commit

Include TDD instructions in every implementer context.

### With requesting-code-review

The two-stage review process IS the code review. For final integration review, use the requesting-code-review skill's review dimensions.

### With systematic-debugging

If a subagent encounters bugs during implementation:
1. Follow systematic-debugging process
2. Find root cause before fixing
3. Write regression test
4. Resume implementation

## Example Workflow

```
[Read plan: docs/plans/auth-feature.md]
[Create todo list with 5 tasks]

--- Task 1: Create User model ---
[Dispatch implementer subagent]
  Implementer: "Should email be unique?"
  You: "Yes, email must be unique"
  Implementer: Implemented, 3/3 tests passing, committed.

[Dispatch spec reviewer]
  Spec reviewer: ✅ PASS — all requirements met

[Dispatch quality reviewer]
  Quality reviewer: ✅ APPROVED — clean code, good tests

[Mark Task 1 complete]

--- Task 2: Password hashing ---
[Dispatch implementer subagent]
  Implementer: No questions, implemented, 5/5 tests passing.

[Dispatch spec reviewer]
  Spec reviewer: ❌ Missing: password strength validation (spec says "min 8 chars")

[Implementer fixes]
  Implementer: Added validation, 7/7 tests passing.

[Dispatch spec reviewer again]
  Spec reviewer: ✅ PASS

[Dispatch quality reviewer]
  Quality reviewer: Important: Magic number 8, extract to constant
  Implementer: Extracted MIN_PASSWORD_LENGTH constant
  Quality reviewer: ✅ APPROVED

[Mark Task 2 complete]

... (continue for all tasks)

[After all tasks: dispatch final integration reviewer]
[Run full test suite: all passing]
[Done!]
```

## Remember

```
Fresh subagent per task
Two-stage review every time
Spec compliance FIRST
Code quality SECOND
Never skip reviews
Catch issues early
```

**Quality is not an accident. It's the result of systematic process.**

## Further reading (load when relevant)

When the orchestration involves significant context usage, long review loops, or complex validation checkpoints, load these references for the specific discipline:

- **`references/context-budget-discipline.md`** — Four-tier context degradation model (PEAK / GOOD / DEGRADING / POOR), read-depth rules that scale with context window size, and early warning signs of silent degradation. Load when a run will clearly consume significant context (multi-phase plans, many subagents, large artifacts).
- **`references/gates-taxonomy.md`** — The four canonical gate types (Pre-flight, Revision, Escalation, Abort) with behavior, recovery, and examples. Load when designing or reviewing any workflow that has validation checkpoints — use the vocabulary explicitly so each gate has defined entry, failure behavior, and resumption rules.

Both references adapted from gsd-build/get-shit-done (MIT © 2025 Lex Christopherson).

© Hermes Agent (adapted from obra/superpowers) · MIT · NousResearch/hermes-agent

Similar skills

All skills for Agent setup: memory, context, orchestration →

subagent-driven-development by other authors comes first.

subagent-driven-development Jesse Vincent The subagent-driven-development skill runs an implementation plan through fresh subagents for independent tasks and reviews each result. Use it for multi-step work in one session. CX Codex +1 237,746installs subagent-driven-development NeoLabHQ subagent-driven-development runs an implementation plan through fresh subagents and puts code review between tasks, so context stays clean and issues surface early. CX Codex +1 1,300installs subagent-delegation chrisbanes The subagent-delegation skill for Claude Code manages subagent task distribution and reassignment. It runs when the user invokes the playbook or explicitly requests delegation. CX Codex +1 123installs test-driven-development Jesse Vincent The test-driven-development skill guides an agent through test-first development: verify a test fails, write the smallest passing change, then refactor. It covers features, bug fixes, and behavior changes. CX Codex +1 259,292installs development-skills hashgraph-online development-skills adds disciplined engineering workflows with risk-tiered execution, on-disk decision chronicles, and fresh-eyes subagent review. CX Codex ★ 1,596repo stars constraint-driven-development addyosmani constraint-driven-development turns vague quality expectations into a written contract for your project. Interviews you about which dimensions matter, then records them in CONSTRAINTS.md as numbers a check can verify. CX Codex +1 49,100installs behavior-driven-development FradSer behavior-driven-development — a BDD skill for feature work, bug fixes, and executable tests. It turns behavior examples into failing tests before implementation. CX Codex +1 151installs doubt-driven-development addyosmani The doubt-driven-development skill sends non-trivial decisions to a fresh-context reviewer looking for flaws. Use it before changes that are hard to undo or costly to debug later. CX Codex +1 72,600installs

From the same set61

The whole set →
document-to-action-items document-to-action-items is a Claude Code skill that extracts obligations, deadlines, and tasks after document extraction, tracing each to its source file and page. CX Codex +1 173installs blocked-page-recovery blocked-page-recovery is a Claude Code skill that searches for copies of pages blocked by 403, 429, paywalls, or bot checks. It recovers research material while keeping source dates clear. CX Codex +1 167installs weekly-review-planning weekly-review-planning — a skill for Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode that runs a structured weekly reset across tasks, calendar, and notes. CX Codex +1 163installs meeting-action-items meeting-action-items — a Claude Code skill that turns meeting notes or a transcript into decisions, owners, and tickets. Run it after a call to draft the follow-up package. CX Codex +1 162installs sdlc-review sdlc-review reviews a task an implementer handed to the Kanban review lane, then approves it, sends it back with findings, or escalates it to a human. CX Codex +1 162installs product-price-monitor product-price-monitor tracks all-in prices for goods, flights, or hotels and triggers alerts upon hitting target thresholds. Filters repeated notifications. CX Codex +1 160installs pixel-art Turn images into era-specific pixel art with NES, Game Boy, or PICO-8 palettes, then add short animated scenes such as rain, snow, or fire. CX Codex +1 138installs box Claude Code skill box manages cloud files, sharing, search, and metadata in Box. CX Codex +1 111installs

Project sponsors

Companies that keep this analytics open