跳转到主要内容
此内容尚未提供您的语言版本,正在以英文显示。

First Principles Framework (FPF) Plugin

插件 活跃

First Principles Framework (FPF) for structured reasoning using workflow command pattern. Implements ADI (Abduction-Deduction-Induction) cycle via propose-hypotheses workflow with fpf-agent for hypothesis generation, logical verification, empirical validation, and auditable decision-making. Includes utility commands for status, query, decay, actualize, and reset.

6 个 Skill 0 个 MCP
目的

To make AI decision-making transparent, auditable, and evidence-based by enforcing a systematic hypothesis-driven reasoning process.

功能

  • Implements FPF ADI cycle for structured reasoning
  • Generates, verifies, and validates competing hypotheses
  • Audits trust scores for evidence-based validation
  • Provides commands for managing decision cycles and knowledge
  • Creates auditable decision rationale records (DRRs)

使用场景

  • Making critical architectural or technical decisions with an auditable trail
  • Preventing premature conclusions by systematically evaluating alternatives
  • Building and leveraging project knowledge over time
  • Ensuring AI-assisted development processes are transparent and accountable

非目标

  • Making unsupervised AI decisions
  • Replacing human judgment in the decision-making process
  • Providing a general-purpose LLM interaction tool

工作流

  1. Initialize FPF context with problem statement
  2. Generate competing hypotheses (Abduction)
  3. Allow user to add hypotheses
  4. Verify logic of each hypothesis (Deduction)
  5. Validate evidence for logically sound hypotheses (Induction)
  6. Audit trust scores for validated hypotheses
  7. Present decision comparison and select winner
  8. Record decision rationale (DRR) and present summary

实践

  • Structured Reasoning
  • Knowledge Management
  • Auditability

Practical Utility

  • warning:Production readinessThe plugin's core functionality, especially its reliance on a large FPF specification (~600k tokens) loaded into a sub-agent with a Sonnet[1m] model, may quickly consume token limits and could hinder practical use in real workflows without careful management.

安装

请先添加 Marketplace

/plugin marketplace add NeoLabHQ/context-engineering-kit
/plugin install fpf@context-engineering-kit

质量评分

88 /100
1 day ago 分析

信任信号

最近提交9 days ago
星标993
许可证GPL-3.0-or-later
状态
查看源代码