agent-fractal-workflow-eval

Audit multi-agent workflow data flow and directive propagation for dead code paths.

Updated Apr 1, 2026

Search Tags:#routing #multi-agent #data-flow #workflow-audit #re-entry #artifact-contracts

Ready to run

npx skills add https://github.com/skurekjakub/docwriter-agent --skill agent-fractal-workflow-eval

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agent-fractal-workflow-eval
Source: https://github.com/skurekjakub/docwriter-agent/tree/main/.github/skills/agent-fractal-workflow-eval
Command: npx skills add https://github.com/skurekjakub/docwriter-agent --skill agent-fractal-workflow-eval

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill helps audit complex multi-agent pipelines to verify prompts, inputs/outputs, and directive propagation across passes, coordinators, and specialists, ensuring no dead code, lost context, or schema drift.

Core Features & Use Cases

Fractal workflow evaluation across orchestrator, coordinators, and specialists to detect dead code paths, misrouted directives, and re-entry gaps.
Data-flow verification and artifact contracts testing to ensure inputs and outputs are consumed appropriately by downstream agents.
Use cases include auditing "audit this workflow" prompts, checking re-entry correctness, and validating routing/document alignment.

Quick Start

Initiate a full audit of a fractal workflow and trace gap-hunter data from gap-analysis through re-entry to specialists, then produce a findings report.

Frequently Asked Questions about agent-fractal-workflow-eval

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

FAQPage Schema

How do I audit multi-agent workflow routing for dead code paths and misrouted directives?▼

To audit multi-agent workflow routing, trace directive propagation and data flow across orchestrators, coordinators, and specialists to detect dead code paths, misrouted directives, and missing artifact contracts. This verifies alignment with architecture documents.

What is a fractal multi-agent workflow evaluation?▼

A fractal multi-agent workflow evaluation assesses end-to-end data flow across nested orchestrators and specialists to verify gap-analysis context, check re-entry correctness, and identify schema drift or lost context preventing proper routing.

How do I verify re-entry context and gap-analysis data flow in a multi-agent pipeline?▼

Verify re-entry context by tracing gap-hunter data through the verification-coordinator down to specialists. This checks whether coordinators read and pass gap-context correctly, ensuring dead branches in mode detection do not prevent gaps from applying.

Does multi-agent workflow auditing check artifact contracts and status schema alignment?▼

Yes, multi-agent workflow auditing verifies artifact contracts by testing that inputs and outputs are consumed appropriately by downstream agents. It also validates status schemas and routing alignment with your architecture documents to prevent data loss.

What causes lost context and schema drift in multi-agent pipelines?▼

Lost context and schema drift in multi-agent pipelines occur when dead branches in mode detection prevent re-entry, coordinators fail to pass gap-context to specialists, or downstream agents consume misrouted directives with missing artifact contracts.