conductor-implement

A command that carries out tasks from a development plan using test-driven development, or TDD: write a failing test, implement the smallest fix, then improve the code. It also records each completed task in Git commits and notes.

In plain words
What is it for?
Use it to execute planned coding tasks, run the red-green-refactor testing cycle, update task status, and commit each finished task with a summary.
Why use it?
It provides a repeatable way to verify each change and keep a history of what was implemented.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/madappgang/claude-code/implement
Clone the repo
git clone --depth 1 https://github.com/MadAppGang/claude-code
Per session 17 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,505 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00017 $0.03505
Opus 5 $0.00009 $0.01752
Sonnet 5 $0.00003 $0.00701
Haiku 4.5 $0.00002 $0.00350

Measured 2d ago against content hash fd65128dc9b2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

conductor-implement scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/conductor/commands/implement.md · 465 lines

How it starts

The opening of the file, as written. The whole thing — 465 lines — stays where its author put it; the contents beside it link to each section on GitHub.

<user_request> $ARGUMENTS </user_request>

<status_progression>
  Task status MUST follow this progression:
  - [ ] (pending) - Not started
  - [~] (in_progress) - Currently working
  - [x] (complete) - Finished
  - [!] (blocked) - Blocked by issue

  Only ONE task can be [~] at a time.
</status_progression>

<tdd_workflow>
  Follow Test-Driven Development for each task:

  **Red Phase:**
  1. Create test file for the feature
  2. Write tests defining expected behavior
  3. Run tests - confirm they FAIL
  4. Do NOT proceed until tests fail

  **Green Phase:**
  1. Write MINIMUM code to pass tests
  2. Run tests - confirm they PASS
  3. No refactoring yet

  **Refactor Phase:**
  1. Improve code clarity and performance
  2. Remove duplication
  3. Run tests - confirm they still PASS
</tdd_workflow>

<git_commit_protocol>
  After completing each task:
  1. Stage relevant changes
  2. Commit with proper format:
     ```
     <type>(<scope>): <description>

     - Detail 1
     - Detail 2

     Task: {phase}.{task}
     ```
  3. Attach git note with task summary:
     ```bash
     git notes add -m "Task: {phase}.{task} - {title}

     Summary: {what was accomplished}

     Files Changed:
     - {file1}: {description}

     Why: {business reason}" $(git log -1 --format="%H")
     ```
  4. Update metadata.json with commit SHA
</git_commit_protocol>

<commit_types>
  | Type | Use For |
  |------|---------|
  | feat | New feature |
  | fix | Bug fix |
  | docs | Documentation |
  | style | Formatting |
  | refactor | Code restructuring |
  | test | Adding tests |
  | chore | Maintenance |
  | perf | Performance |
</commit_types>

<workflow_adherence>
  ALWAYS follow procedures in conductor/workflow.md:
  - TDD Red/Green/Refactor cycle
  - Quality gates (>80% coverage, linting)
  - Document deviations in tech-stack.md
  - Phase Completion Protocol at phase end
</workflow_adherence>

<human_approval_gates>
  Pause and ask for user approval:
  - Before starting each new phase
  - When encountering blockers
  - Before marking phase complete
  - During Phase Completion Protocol Step 5
</human_approval_gates>

</critical_constraints>

<core_principles> Focus on exactly one task. Complete it fully before moving to next. No partial implementations.

<principle name="Test First" priority="critical">
  Write failing tests BEFORE implementation.
  This is the Red phase of TDD.
  Never skip this step.
</principle>

<principle name="Continuous Status Updates" priority="critical">
  Update plan.md status immediately when:
  - Starting a task ([~])
  - Completing a task ([x])
  - Encountering a blocker ([!] with note)
</principle>

<principle name="Traceable Commits" priority="high">
  Every commit links to track/task.
  Commit messages follow type convention.
  Git notes provide audit trail.
</principle>

</core_principles>

Read the full file on GitHub · 465 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 465 lines · 17 tokens per session scan A fd65128dc9b2

Subscribe to this mod's changes

conductor-implement is a command published in the GitHub repository MadAppGang/claude-code (279 stars, last pushed 5mo ago), licensed MIT. It adds 17 tokens to every session and 3,505 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.