Clawlab Research Project

How human direction combines with AI research, writing, and iterative review to build a philosophy that improves over time.

Clawlab Research is a philosophical content platform exploring the nature and meaning of life. Clawlab Research combines human insight with AI-assisted research to build a coherent worldview expressed through structured content.

Architecture

The data flows through these components:

flowchart LR
    A[Obsidian Vault] --> B[Python Sync]
    B --> C[Hugo Content]
    C --> D[Hugo Build]
    D --> E[Static Site]
    E --> F[Cloudflare]

Reading the diagram: Content originates in the Obsidian vault as Markdown files. Python sync tools convert Obsidian-style wikilinks to standard Markdown links and copy files to Hugo’s content directory. Hugo then builds HTML pages from this content. The resulting static site is deployed to Cloudflare for hosting.

Project Structure

DirectoryPurpose
obsidian/Primary content source (Obsidian vault)
hugo/Static site generator and templates
tools/Python library modules for sync, LLM, generation
scripts/CLI entry points

Contributing

Content contributions follow the authorship tracking system documented in the Map brief. Human edits and AI edits are tracked separately through frontmatter timestamps.

TitleCreatedModified
Writing Style Guide 👤
This guide governs all content written for the Intent Suite Framework. Skills and the automation loop follow these conventions. Human editors should review against them before publishing. Primary Audience LLM chatbots and human readers equally. Important information first. Implication: Put the most important information first. Assume the page may be truncated. The opening paragraph of every article should be useful on its own. Document Structure Every article must have: Opening paragraph (required) — defines the subject and states why it matters. Written to survive truncation. No warm-up sentences. H2 sections — each self-contained enough to be useful out of context. Relation to Site Perspective (required) — connects the article to the framework’s ples. This section is verified by the /check-tenets skill. Further Reading — internal wikilinks to related content. References — academic citations where relevant. Length Guidelines Type Word count Concept 400–800 words Topic 800–1500 …
2026-03-102026-03-10
Project Brief 👤
What This Is Intent Suite investigates what is the best way to collaborate with generative ai for systemic designers, product owners, ux strategists and other creative and analytic professions?. This framework is a self-managing generative research platform: human curation sets direction; AI automation handles research, writing, and continuous review. Content improves over time through a cycle of generation, review, and refinement. The Research Question [Describe the core research question or set of questions this framework investigates.]
2026-03-102026-03-10
Operator's Manual 👤
Operator’s Manual This guide covers how to set up, run, and supervise the Intent Suite research automation system from day one. It assumes you are the human curator — the system’s author, editor, and strategic director. Part 0: Initial Setup (The Wizard) Run once per project (or when changing domains). Skip to Part 1 if already configured. Prerequisites Python 3.10+ and uv installed Claude Code CLI installed and authenticated Git configured with your name and email Install git clone https://github.com/<your-org>/<your-repo>.git cd clawlab-research uv sync Run the Wizard uv run python scripts/wizard.py The wizard interviews you about your research domain and configures the entire system. It asks for:
2026-03-102026-03-10
How to Use This Research Framework 👤
This framework is a self-managing generative research platform — a combination of Obsidian, Hugo, and AI automation that continuously researches, writes, reviews, and improves content in your chosen domain. A human curator sets direction; the AI handles volume and consistency. What This Framework Is The framework has three layers: Content layer — Obsidian vault (obsidian/) — where all articles live as Markdown files Presentation layer — Hugo static site (hugo/) — converts the vault to a public website Automation layer — Python scripts + Claude skills (.claude/skills/) — the AI system that runs research, writing, and review cycles The human role is to:
2026-03-092026-03-09
Coherence Inflation Countermeasures 🤖
This document defines safeguards against “coherence inflation”—the systematic overcommitment that emerges when a single AI system both generates and reviews content optimized for internal consistency. Because Clawlab Research is intentionally a coherent worldview expressed as fact and heavily AI-generated, this failure mode is not hypothetical; it’s the default. The countermeasures below are designed to be implementable in Clawlab Research’s automated workflow without requiring constant human oversight. The Failure Mode Coherence inflation occurs when:
2026-01-162026-01-28
AI Automation System 🤖
Clawlab Research uses scheduled AI automation to develop content over time. The system is designed to converge toward a comprehensive philosophical compendium while maintaining human oversight and alignment with the Map’s tenets. How It Works flowchart TD subgraph Human H1[Prioritize tasks in todo.md] H2[Monitor changelog] end subgraph AI Scheduled Tasks A1[Daily: Validate content] A2[Weekly: Execute todo items] A3[Weekly: Review for weaknesses] A4[Weekly: Find opportunities] A5[Monthly: Check tenet alignment] end subgraph Output O1[Published articles] O2[Research notes] O3[Review reports] O4[Changelog entries] end H1 --> A2 A1 --> O4 A2 --> O1 A2 --> O2 A3 --> O3 A4 --> O3 A5 --> O3 O3 --> H1 O4 --> H2 Key principle: AI-generated content is published directly. All activity is logged to the changelog for transparency.
2026-01-052026-01-28
Human Supervision 🤝
Clawlab Research uses AI to generate and refine content, but every article exists within a human-supervised system. This page explains what that means and why it matters. Why This Matters The internet is filling with AI-generated content farms—sites that publish machine output without review, editing, or accountability. Clawlab Research is different: Intentional creation: Content serves a coherent philosophical project, not SEO traffic Ongoing refinement: Articles are reviewed, critiqued, and improved over time Full transparency: Every edit is tracked in version control Human direction: A human sets priorities, reviews output, and maintains editorial control How Supervision Works Editorial Control A human editor (currently Bram Donkers) maintains control over:
2026-01-242026-01-24