How do I visualize dependencies using this skill?

When you query a specific hypothesis ID, the skill automatically generates a text-based tree showing all 'depends_on' relationships, including their individual reliability scores.

What is an FPF knowledge base?

A Formalized Proposal Framework (FPF) knowledge base is a structured repository of technical hypotheses and design decisions categorized into layers (L0 to L2) based on their verification status.

Can I see why a technical decision was rejected?

Yes, by querying Design Rationale Records (DRR) or searching the 'invalid' knowledge layer, you can retrieve the context and reasoning behind rejected hypotheses.

How does the R_eff score work in search results?

R_eff represents the Effective Reliability of a hypothesis. The skill calculates this by evaluating evidence, audit sections, and applying penalties for external documentation or weak dependencies.

FPF Knowledge Query

Name: FPF Knowledge Query
Author: NeoLabHQ

byNeoLabHQ

•

687

•

学习与文档

Searches the FPF knowledge base to retrieve architectural hypotheses, assurance data, and design rationale records.

The FPF Knowledge Query skill enables Claude to interface directly with the Formalized Proposal Framework (FPF) knowledge base, providing instant access to technical hypotheses and architectural decisions. It allows users to query system designs by maturity layer (L0-L2), inspect effective reliability scores (R_eff), visualize complex dependency graphs, and review Design Rationale Records (DRR). This tool is essential for developers and architects who need to validate technical choices against existing project evidence and maintain structural integrity across complex codebases.

主要功能

01Automated visualization of hypothesis dependency trees with congruence levels

02Detailed R_eff (Effective Reliability) breakdowns including weakest link analysis

03687 GitHub stars

04Direct retrieval of Design Rationale Records (DRR) for decision auditing

05Multi-layered search across proposed, verified, validated, and rejected hypotheses

06Context-aware filtering by keyword, specific ID, or knowledge maturity layer

使用场景

01Onboarding new team members by providing searchable access to historical architectural decisions

02Identifying technical risks by visualizing deep dependency chains and reliability penalties

03Auditing the reliability and evidence status of specific system components before implementation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add neolabhq/context-engineering-kit query

For use in Claude.ai and ChatGPT

Download Skill