knowledge-acquisition

Knowledge is not gathered — it is systematically excavated. The difference between reading papers and acquiring knowledge is the same as between tourism and cartography.

🧠 Knowledge Acquisition

Systematic Research Knowledge Acquisition Engine for Academic Research

🧭 Part of the De-Anthropocentric Research Engine. This repository is one of nine composable research packages that make up DARE — the full autonomous research-orchestration system. DARE bundles this package together with the others into a single self-contained clone, unified under one orchestrator. To use these skills as intended — with the spec-driven orchestrator and cross-package routing — clone the main repository rather than this repo alone.

Knowledge Acquisition is not a literature search tool. It is a complete research intelligence pipeline. You provide a research intent — it surveys literature, mines patents, audits benchmarks, synthesizes cross-study evidence, and establishes SOTA baselines. Five autonomous campaigns, each a self-contained research activity domain with quantitative budget enforcement.

⚡ What It Does

• 📚 Literature Survey — autonomous literature survey across 5 paradigms (scoping, systematic, deep, narrative, snowball). Searches, screens, reads, categorizes, identifies gaps
• ⚖️ Patent Mining — patent landscape analysis with prior art search, claim decomposition, competitive intelligence, white-space mapping, and trend tracking
• 🏋️ Benchmark Archaeology — systematic excavation of AI/ML evaluation methodology. Audits benchmark quality, detects saturation, probes construct validity, maps coverage gaps
• 📊 Meta-Analysis — cross-study statistical synthesis with effect size extraction, heterogeneity investigation, publication bias detection, and GRADE evidence assessment
• 🎯 Baseline Establishment — SOTA performance baseline collection. Method inventory, performance extraction, condition standardization, discrepancy analysis, progress quantification

🎯 Design Philosophy

🎖️ Campaign-Based Architecture

Each campaign is a self-contained research activity domain. The entry point routes to the correct campaign based on user intent, then the campaign autonomously selects strategies and executes with quantitative budget enforcement.

📐 Four-Level Hierarchy

ENTRY.md (root)
  → Campaign (5): self-contained research activity domain
    → Strategy: selected by analysis purpose/intent
      → Tactic: multi-step orchestration pattern (reusable across strategies)
        → SOP: single operation (import or subagent)

📏 Budget Enforcement

Every strategy embeds quantitative floors using domain-natural units (papers read, patents analyzed, benchmarks audited, effect sizes extracted). State Ledgers track progress. Budget gates prevent premature exit.

🏗️ Architecture

┌───────────────────────────────────────────────────────────────┐
│  ENTRY POINT (ENTRY.md)                                       │
│  Routes to campaign based on research intent                  │
├───────────────────────────────────────────────────────────────┤
│  CAMPAIGN (5)                                                 │
│  literature-survey, patent-mining, benchmark-archaeology,     │
│  meta-analysis, baseline-establishment                        │
├───────────────────────────────────────────────────────────────┤
│  STRATEGY (25)                                                │
│  5 per campaign, selected by analysis purpose                 │
├───────────────────────────────────────────────────────────────┤
│  TACTIC (15)                                                  │
│  3 per campaign, multi-step orchestration patterns            │
├───────────────────────────────────────────────────────────────┤
│  SOP (55)                                                     │
│  Import (5 shared) + Subagent (~50)                           │
├───────────────────────────────────────────────────────────────┤
│  EXTERNAL SKILLS + MCP TOOLS                                  │
│  web-browsing, literature-engine, subagent-spawning,          │
│  context-management, brave-search, apify, alphaxiv,           │
│  semantic-scholar                                             │
└───────────────────────────────────────────────────────────────┘

🧩 Campaign Inventory

📁 Repository Structure

knowledge-acquisition/
├── ENTRY.md                          # Root entry point — campaign routing
├── skills/
│   ├── literature-survey/            # Campaign
│   ├── patent-mining/                # Campaign
│   ├── benchmark-archaeology/        # Campaign
│   ├── meta-analysis/                # Campaign
│   ├── baseline-establishment/       # Campaign
│   ├── scoping-survey/              # Strategy (literature-survey)
│   ├── systematic-survey/           # Strategy (literature-survey)
│   ├── ...                          # 95 more skill directories
│   └── web-search/                  # Import SOP (shared)
├── tests/
│   └── integration-prompt.md
└── README.md

All skills live flat in skills/. Organization is via frontmatter used-by field, not directory nesting.

🔌 Dependencies

🔧 MCP Servers Required

🚀 Quick Start

1. Clone:

git clone https://github.com/yogsoth-ai/knowledge-acquisition.git

2. Ensure sibling skill repos are available: web-browsing, literature-engine, subagent-spawning, context-management

3. Configure MCP servers (brave-search, apify, alphaxiv, semantic-scholar) in your Claude Code session.

4. Invoke:

/knowledge-acquisition I need to understand the patent landscape for protein language models

The engine routes to the correct campaign (patent-mining in this case) and executes autonomously.

📄 License

Apache-2.0

A component of the De-Anthropocentric Research Engine, part of the Yogsoth AI ecosystem. Built by Pthahnix.