Can I use different AI models for different tasks?

Yes, the skill automatically selects models based on whether a task is mechanical, integration-focused, or architectural in nature.

Is there a final review of the generated code?

Yes, after all sequential tasks in the plan are completed, the skill automatically triggers the dodi-dev:review process for a final check.

How does it verify that a task is actually finished?

It manually checks git diffs, logs, and test outputs to confirm changes were successfully applied and requirements were met.

Does the implement skill support parallel execution?

No, it processes tasks sequentially to prevent file conflicts and ensure a clean, understandable Git history.

What happens if a subagent gets stuck or lacks information?

The skill identifies 'BLOCKED' or 'NEEDS_CONTEXT' statuses and will either ask the user for more info or suggest using a more capable model.

Plan Implementation Agent

Name: Plan Implementation Agent
Author: dodi-hq

bydodi-hq

0•

Productividad y Flujo de Trabajo

Executes complex implementation plans by dispatching specialized subagents to handle individual tasks with automated verification and commits.

The dodi-dev implement skill automates the transition from planning to execution by managing the full lifecycle of development tasks. It sequentially dispatches autonomous subagents to handle specific items in a plan, ensuring each change is verified through git diffs and test results before moving to the next. By intelligently selecting models based on task complexity—from mechanical file updates to deep architectural changes—it maintains a high velocity while ensuring code quality and repository integrity through a structured self-review and commit process.

Características Principales

010 GitHub stars

02Integrated git commit management for every completed task

03Sequential subagent dispatching for atomic task execution

04Intelligent model selection based on task complexity

05Structured status handling for blocked tasks and context needs

06Automated verification of git diffs and test results

Casos de Uso

01Ensuring consistent test-driven development by enforcing verification gates

02Executing multi-step feature implementation plans without manual oversight

03Automating large-scale refactoring across multiple files and directories

Características Principales

010 GitHub stars

02Integrated git commit management for every completed task

03Sequential subagent dispatching for atomic task execution

04Intelligent model selection based on task complexity

05Structured status handling for blocked tasks and context needs

06Automated verification of git diffs and test results

Casos de Uso

01Ensuring consistent test-driven development by enforcing verification gates

02Executing multi-step feature implementation plans without manual oversight

03Automating large-scale refactoring across multiple files and directories