GitHub Repository

Polywave Agent Skill for Claude Code: parallel AI agents with disjoint file ownership, worktree isolation, and human-reviewed plans. Requires polywave-tools CLI.

2 starsShell

Scout-and-Wave – Parallel agent coordination via prompts

Name: Scout-and-Wave – Parallel agent coordination via prompts
Availability: InStock
Author: daynablackwell

by daynablackwell·Mar 4, 2026·2 points·0 comments

Visit Project View on HN

AI Analysis

●●●BangerBig BrainWizardryZero to One

Multi-agent parallelism solved via markdown prompts, not framework magic—genuinely novel.

Strengths

•Core insight is elegant: prevent merge conflicts at planning time (not merge time) via invariants.
•Entire protocol lives in natural language—no binary, no framework coupling, Claude Code native.
•Structured human checkpoints (suitability gate, IMPL review) avoid runaway agent behavior by design.

Weaknesses

•Requires Claude Code setup and willingness to use git worktrees—not portable to other LLM platforms.
•No public benchmark or case study showing >2 agents actually working without conflicts on real projects.

Post Description

I built a coordination protocol that lets multiple Claude Code agents work on the same codebase in parallel without merge conflicts. The protocol is encoded entirely in markdown prompts, no binary or framework.

The core idea: prevent conflicts at planning time through two invariants: 1. Disjoint file ownership: no two agents in the same wave touch the same file 2. Frozen interface contracts: agents code against exact type signatures defined upfront

How it works:

- Scout agent analyzes the codebase, runs a suitability gate (5 questions), produces a coordination artifact (IMPL doc) with file ownership table and interface contracts - Human reviews and approves - Scaffold agent creates shared type files from approved contracts, compiles them - Wave agents implement in parallel, each in its own git worktree - Orchestrator merges sequentially, conflict-free by construction

The interesting part: the entire protocol lives in natural language. Invariants are embedded verbatim in the prompts at their enforcement points. A capable LLM follows them consistently. This proves you can encode coordination protocols in prompts and get structural safety guarantees, which has implications beyond parallel coding.

Repo: https://github.com/blackwell-systems/scout-and-wave Formal spec with invariants I1-I6: https://github.com/blackwell-systems/scout-and-wave/blob/mai...

I built this because I kept spawning multiple Claude Code sessions in separate terminals and having them conflict. Worktrees isolated execution but didn't prevent the conflicts. Realized the missing piece was coordination before execution, not better infrastructure.

Tradeoffs: rigidity (if an interface is wrong mid-wave, you restart), LLM drift risk (prompts can be reasoned around, unlike code enforcement), and currently Claude Code specific (though the protocol is portable).

Dogfooded on 50+ features across multiple projects. Failure rate: suitability gate rejects ~20-30% of requests (correct rejections), ~85-90% of suitable decompositions complete on first try.