wave-aggregator

Merge Wave A and Wave B outputs into docs/project-context.yaml.

Updated Feb 6, 2026
One-click install
npx skills add https://github.com/sizukutamago/blueprint-plugin --skill wave-aggregator
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wave-aggregator
Source: https://github.com/sizukutamago/blueprint-plugin/tree/main/skills/wave-aggregator
Command: npx skills add https://github.com/sizukutamago/blueprint-plugin --skill wave-aggregator

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill consolidates parallel Wave A and Wave B outputs into a single, authoritative Blackboard file, eliminating conflicts and ensuring consistency in design decisions.

Core Features & Use Cases

  • Unified context: merges outputs from Wave A/B and updates docs/project-context.yaml.
  • Conflict resolution: performs a two-step reduce (normalization + adjudication) to resolve contradictions.
  • Deterministic writes: enforces the single-writer rule for the blackboard and reports unresolved issues to the Lead.

Quick Start

Run the wave-aggregator once Wave A and Wave B finish to merge their outputs and update docs/project-context.yaml with the unified context.

Frequently Asked Questions about wave-aggregator

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

FAQPage Schema
How do I merge conflicting parallel outputs from agent teams into a unified context?

To merge conflicting parallel outputs into a unified context, aggregate Wave A and Wave B results using a two-step reduce process of normalization and adjudication to resolve contradictions, then write the coherent state to docs/project-context.yaml.

What is the best way to resolve conflicting decisions produced by parallel wave phases?

Resolving conflicting decisions from parallel wave phases requires an adjudication pass driven by contract_outputs from the Lead, applying normalization to align Wave A/B outputs before enforcing a deterministic single-writer update to the blackboard.

Do I need contract_outputs from the Lead to perform blackboard conflict resolution?

Yes, you need contract_outputs from the Lead to drive the adjudication pass for blackboard conflict resolution, ensuring that contradictions between Wave A and Wave B are resolved authoritatively and unresolved issues are properly reported.

How does a single-writer rule work when updating a shared blackboard file?

A single-writer rule ensures deterministic writes to the shared blackboard file by restricting updates to a single aggregation pass, preventing race conditions and maintaining consistency when merging parallel agent team outputs into docs/project-context.yaml.

What happens to unresolved issues during parallel wave aggregation?

During parallel wave aggregation, unresolved issues that cannot be reconciled through normalization and adjudication are reported back to the Lead, ensuring the unified context written to the blackboard remains authoritative and free of silent failures.

Can I use wave-aggregator to consolidate outputs without a docs/project-context.yaml file?

No, consolidating parallel wave outputs requires docs/project-context.yaml as the deterministic write target, ensuring the unified context is persisted to a single authoritative blackboard file structure.