next-move

Predict a five-node meta-DAG of high-impact skills from git state and project context.

10|1|Updated Mar 17, 2026
One-click install
npx skills add https://github.com/curiositech/windags-skills --skill next-move
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: next-move
Source: https://github.com/curiositech/windags-skills/tree/main/skills/next-move
Command: npx skills add https://github.com/curiositech/windags-skills --skill next-move

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Analyzes a project's current context to produce a predicted five-node meta-DAG of the highest-impact skills to run next, enabling deterministic planning and prioritization.

Core Features & Use Cases

  • Simulates a five-agent pipeline (Sensemaker → Decomposer → Skill Selector + PreMortem → Synthesizer) to output a ready-to-execute PredictedDAG.
  • Provides wave-based task sequencing with explicit input/output contracts and risk signals for proactive decision making.
  • Works from project context (git state, CLAUDE.md, recent commits, and workspace metadata) to align actions with goals and constraints.

Quick Start

Provide the project context to the predictor and request a PredictedDAG for the current sprint.

Frequently Asked Questions about next-move

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

FAQPage Schema
How do I predict the next high-impact tasks for my current sprint based on project context?

To predict high-impact tasks for a sprint, this skill analyzes project context like git state and recent commits to generate a five-node meta-DAG. It outputs a structured PredictedDAG with wave definitions and a premortem summary for deterministic planning and prioritization.

What is a meta-DAG for task sequencing and how does it improve project planning?

A meta-DAG for task sequencing is a directed acyclic graph that maps high-impact skills into a five-wave execution pipeline. It improves project planning by establishing deterministic input/output contracts between nodes and integrating risk analysis for proactive decision making.

Can I use git state and recent commits to automate risk analysis for upcoming project work?

Yes, you can use git state, branch names, and recent commits to automate risk analysis. The skill evaluates this project metadata to assemble a five-wave DAG and generates a premortem summary, identifying potential risks before execution begins.

How do I sequence project tasks into waves with clear input and output contracts?

Sequencing tasks into waves with clear input/output contracts is handled by simulating a five-agent pipeline including a Decomposer and Skill Selector. This process analyzes your workspace metadata to structure a deterministic five-node DAG ready for execution.

Does generating a predicted DAG require specific project metadata or configuration files?

Generating a predicted DAG requires project metadata such as git state, branch name, recent commits, and CLAUDE.md contents. The predictor analyzes these inputs to align actions with your project goals and constraints without needing external dependencies.

When should I use a DAG-based approach over manual prioritization for complex projects?

Use a DAG-based approach over manual prioritization when working with complex projects requiring deterministic planning and risk-aware decision making. The five-wave structure provides explicit risk signals and input/output contracts that manual task sequencing often misses.