Build teams of AI agents that collaborate, automate workflows, and complete complex tasks together.
Ordewell
0Popularity
About Ordewell
Ordewell is a task orchestration tool designed for coding agents. It generates an ordered plan of agent tasks by reading the repository and clarifying vague goals before execution. Each task is executed in a separate agent session with specified runners, models, thinking effort, and modes. The tool verifies task completion using unique markers in the output and maintains a dependency graph to ensure correct execution order. It supports parallel execution of independent tasks and provides a VS Code extension for planning, editing, and monitoring tasks within the editor. The planner operates in read-only mode and cannot modify the repository, while runners handle actual code changes. Users can pause tasks before irreversible steps and approve or reject decisions based on gathered evidence. The tool also offers a terminal UI for remote use over SSH and a command-line interface for pipeline integration.
Key features
- Plan generation with task ordering and dependency graph
- VS Code extension for integrated task management
- Terminal UI for remote SSH use
- Command-line interface for pipeline integration
- Task verification via completion markers and exit codes
- Parallel execution of independent tasks
- Editable plans before execution
- Support for multiple coding agents (Claude Code, Codex, OpenCode)
Use cases
- Orchestrating multi-step coding tasks across a repository
- Planning and executing API changes with rate limiting
- Managing security refactoring tasks with appropriate models
Pros
- Generates ordered task plans with specified runners, models, and dependencies
- Supports multiple coding agents (Claude Code, Codex, OpenCode)
- Provides VS Code extension and terminal UI for local and remote use
- Allows editing plans before execution without round-trips to AI
- Verifies task completion using exit codes and completion markers
Cons
- Requires tmux on all platforms and WSL on Windows
- Planner operates in read-only mode; mutation handled by runners only
Frequently asked questions about Ordewell
What does Ordewell do?
Ordewell generates an ordered plan of agent tasks by reading the repository and clarifying vague goals before execution. It executes each task in a separate agent session with specified runners, models, thinking effort, and modes, verifying completion using unique markers in the output.
Who is Ordewell designed for?
Ordewell is designed for developers and teams working with coding agents who need structured task planning, execution oversight, and verification. It suits users who want to orchestrate AI-driven code changes with clear dependencies and evidence-based completion.
How does Ordewell verify task completion?
Ordewell verifies task completion by checking for a unique completion marker in the runner's output. The exit code is retained as evidence, and the model does not influence the verdict.
Can Ordewell modify the repository?
No, the planner operates in read-only mode and cannot modify the repository. Runners handle actual code changes, while the planner only generates and manages the task plan.
Does Ordewell support parallel execution of tasks?
Yes, Ordewell supports parallel execution of independent tasks while maintaining a dependency graph to ensure correct execution order.
How do I get started with Ordewell?
Users can get started with Ordewell by installing it via Node.js on macOS, Linux, or Windows, then using the VS Code extension or terminal UI to plan, edit, and execute tasks. The tool can also be integrated into pipelines via its command-line interface.
