parallel-dispatch-dag

Build a dependency DAG from scope cards and compute level-based parallel dispatch order.

150|48|Updated Jan 24, 2026
One-click install
npx skills add https://github.com/irahardianto/awesome-agv --skill parallel-dispatch-dag
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: parallel-dispatch-dag
Source: https://github.com/irahardianto/awesome-agv/tree/main/.agents/skills/parallel-dispatch-dag
Command: npx skills add https://github.com/irahardianto/awesome-agv --skill parallel-dispatch-dag

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents agents from executing interdependent scope tasks out of order by constructing and validating a dependency DAG so each parallel “level” runs only after its required predecessors complete.

Core Features & Use Cases

  • DAG construction from scope cards: Converts each scope card into a node (with agent type, scope, phase, and type) and turns hard dependencies into directed edges.
  • Topological level scheduling: Performs a topological sort and groups nodes into dependency “levels” so all nodes in the same level can run concurrently.
  • Safety validation before execution: Detects cycles, blocks dangling references, enforces MECE write-scope boundaries via the ownership skill, and applies phase ordering constraints (e.g., DESIGN before BUILD).
  • Failure-aware dynamic re-planning: Invalidates downstream dependents on failure, re-decomposes only the affected sub-tree, and resumes from the correct replanned level.
  • Execution protocol with merge points: Creates git worktrees for write-nodes, dispatches per level, merges after each level, and continues through phase transitions according to orchestrate.toml rules.

Quick Start

Use parallel-dispatch-dag after parallel-dispatch-decomposition generates scope cards, so the system builds a validated dependency graph and dispatches independent scopes in parallel level-by-level.

Frequently Asked Questions about parallel-dispatch-dag

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

FAQPage Schema
How do I schedule parallel agent workflows without dependency conflicts?

Parallel agent workflows are scheduled by building a directed acyclic graph from scope cards and grouping tasks into topological levels, ensuring each level runs concurrently only after required predecessors complete.

What is topological level dispatch for agentic workflows?

Topological level dispatch is a scheduling method that performs a topological sort on dependency graphs to group independent scope cards into parallel execution levels, allowing safe concurrent execution across phases like SCOUT, DESIGN, and BUILD.

How do I validate cycles and dangling dependencies in a workflow DAG?

Workflow DAG validation detects cycles and blocks dangling references before execution by analyzing directed edges between scope cards, enforcing phase ordering constraints, and checking MECE write-scope boundaries to ensure deterministic execution.

Can I use git worktrees to coordinate parallel write scopes across phases?

Git worktrees coordinate parallel write scopes by creating isolated worktrees for write-nodes, dispatching tasks per dependency level, and merging results after each level completes to maintain safe execution boundaries.

What happens to downstream dependent tasks when a parallel scope fails?

When a parallel scope fails, the system invalidates downstream dependents, dynamically re-decomposes only the affected sub-tree, and resumes execution from the correct replanned dependency level.

Do I need scope cards before dispatching tasks in a DAG?

Yes, scope cards are required as input. The DAG builder converts each scope card into a node with agent type, scope, phase, and type, turning hard dependencies into directed edges for parallel dispatch scheduling.