aod-orchestrate

Translate /aod.blueprint into priority-ordered waves of tasks and batches.

86|21|Updated Mar 21, 2026
One-click install
npx skills add https://github.com/davidmatousek/tachi --skill aod-orchestrate
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aod-orchestrate
Source: https://github.com/davidmatousek/tachi/tree/main/.claude/skills/aod-orchestrate
Command: npx skills add https://github.com/davidmatousek/tachi --skill aod-orchestrate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrates multi-feature execution by translating a /aod.blueprint into priority-ordered waves of tasks, batches, and monitored sessions, enabling scalable and governed deployment of features across a project.

Core Features & Use Cases

  • Priority-based wave planning: groups issues into P0/P1/P2 waves with dependency awareness.
  • Automated task and batch management: creates Task records, spawns batches, and tracks sessions with per-wave progression.
  • Governance and recovery: supports --dry-run and --yes, checkpointing, crash recovery via state files, and abort/retry workflows.

Quick Start

Invoke the aod.orchestrate skill to orchestrate features from the current blueprint in priority-ordered waves.

Frequently Asked Questions about aod-orchestrate

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I orchestrate multi-feature execution with priority-ordered waves?

Multi-feature execution is orchestrated by translating an aod.blueprint into priority-ordered waves of tasks and batches. The system groups issues into P0, P1, and P2 waves with dependency awareness to dispatch features sequentially.

How does dependency checking work before dispatching a wave of tasks?

Dependency checking works by running pre-wave checks before creating tasks and spawning batches. This ensures that all required dependencies are resolved for the specific priority wave before any monitored session execution begins.

Can I use dry-run mode for governance-aware workflow deployments?

Yes, you can use dry-run mode for governance-aware deployments by passing the --dry-run option. This allows you to validate wave planning and task creation without executing the actual changes.

What is the best way to recover a crashed batch processing workflow?

The best way to recover a crashed batch processing workflow is through state files and checkpointing. The orchestration system supports crash recovery to resume per-wave polling and task progression from the last saved state.

Does task management integrate with GitHub metadata for issue selection?

Yes, task management integrates with the AOD API and GitHub metadata for configurable issue selection. This allows the orchestration process to automatically pull feature definitions and translate them into tracked task records.

When should I use auto-confirm options for sequential feature dispatching?

You should use the --yes auto-confirm option for sequential feature dispatching when executing fully automated, governed pipelines. It skips manual approval prompts between waves to streamline continuous deployment workflows.