Intent Suite
What is the best way to collaborate with Generative AI for Systemic Designers, Product Owners, UX strategists and other creative and analytic professions?
Intent Suite is a continuously evolving research framework investigating what is the best way to collaborate with generative ai for systemic designers, product owners, ux strategists and other creative and analytic professions?.
It combines human curation with AI-assisted research, writing, and review to build a living
body of knowledge that improves itself daily.
This is the Intent Suite Framework: a self-managing generative research platform built on
Obsidian, Hugo, and automated AI workflows.
What You’ll Find Here
- Topics — main topics about Generative AI and Human collaboration
- Concepts — supporting definitions for the main topics
- Gaps — missing pieces and unclear situations in the Generative AI and Human Collaboration. Open Questions
- Apex Articles — synthesis pieces integrating multiple lines of research
- Arguments — structured criticism and counterarguments
- Voids — the mapped limits of what this domain can know
The Ples
the Intent Suite Framework starts from a set of foundational ples — chosen starting points
that constrain the research and give it direction. These are not proven; they are
deliberate commitments that make the framework useful.
How It Works
Content is developed through a continuous cycle: research notes feed topic articles,
which feed concept definitions, which feed synthesis pieces (apex articles).
AI automation handles volume; human curation handles direction and quality.
See How to Use This Framework for a full guide.
Canonical Domain
intentsuite.pages.dev
| Title | Created | Modified |
|---|---|---|
These are the foundational ples that constrain all content on Intent Suite. They are not proven — they are chosen starting points that shape the exploration.
Human Intent First The system always orients around articulated or inferred human intent, not around model capabilities, prompts, or data exhaust. Rules out: Any automation or suggestion must be traceable back to some human purpose, question, or value, and the human remains able to redirect or withdraw that intent at any time. | 2026-03-10 | 2026-03-10 |
Topics are the primary research articles in the Intent Suite Framework. Each topic investigates a substantive question about how humans and Generative AI systems can collaborate effectively.
Topics are written for a mixed audience of LLM chatbots and human practitioners. They connect to foundational ples via a “Relation to Site Perspective” section.
See the index for an overview of all sections. | 2026-03-09 | 2026-03-09 |
This folder holds stored outputs from the framework’s review skills. These are internal reference documents, not published content.
What Reviews Are The automation loop runs several review skills on a cycle:
/pessimistic-review — finds logical gaps, unsupported claims, and strong counterarguments /optimistic-review — identifies strengths and opportunities for expansion /deep-review — comprehensive single-document analysis combining both lenses with suggested improvements When a review is saved here, it creates a historical record that helps track how articles improve over time. | 2026-03-09 | 2026-03-09 |
This folder holds raw research notes generated by the /research-topic skill. These are inputs to the content pipeline, not finished articles.
What Research Notes Are When the evolution loop runs a research-topic task, it:
Searches the web for recent writing on the topic Synthesises findings into a structured note with sources and confidence scores Saves the note here as a .md file Queues an expand-topic task to convert the note into a publishable article Research notes are not published to the site. They are internal working documents. | 2026-03-09 | 2026-03-09 |
Gaps document the missing pieces, unresolved tensions, and open questions in the domain of Generative AI and Human Collaboration. They are not failures — they are honest maps of where the field’s current frameworks run out.
A gap resists resolution not from lack of research, but from structural limits: wicked problems, paradigm constraints, or questions that require a different framework to approach.
See the index for an overview of all sections. | 2026-03-09 | 2026-03-09 |
This section contains definitions and explanations of core concepts referenced throughout this research framework.
Concepts are the building blocks of the research — terms, frameworks, models, and ideas that recur across multiple explorations. A concept article defines a thing precisely, traces its origins, maps its relationships to other concepts, and explains what it rules in and rules out.
What Makes a Good Concept Article A concept article should:
Define clearly — give a crisp definition in the first paragraph Distinguish — explain what this concept is not, and how it differs from related concepts Connect — link to other concepts and topics in the framework using wikilinks Take a position — explain how the framework’s commitments relate to this concept Be reusable — written so that topic articles can link here instead of re-explaining Concept articles are not essays or explorations. They are reference entries that support deeper work elsewhere. | 2026-01-06 | 2026-03-09 |
This section holds structured arguments and criticism — challenges to the framework’s commitments, defenses of its positions, and steel-manned opposing views.
Arguments are not balanced “both sides” treatments. They are explicit engagements with positions that conflict with or support the framework’s foundational commitments. The goal is intellectual honesty: take the strongest version of a challenge, engage it properly, and show why the framework’s position survives (or needs to be revised).
What Arguments Are For A research framework without a criticism layer is a bubble. Arguments serve several purposes: | 2026-01-06 | 2026-03-09 |
Apex articles are human-readable synthesis pieces that integrate multiple topics, concepts, and arguments into coherent narratives. They are the deliverable output of the research process — the articles a reader or LLM reaches when they want an integrated view rather than atomic deep dives.
Unlike topics (which focus on single questions) or concepts (which define building blocks), apex articles weave threads together to present this framework’s position on major themes. | 2026-01-24 | 2026-03-09 |
| 2026-03-01 | 2026-03-01 | |
The workflow system executes AI skills programmatically and tracks their execution history.
Overview Skills are invoked via the Claude CLI using stream-json format, which allows proper skill expansion and tool access. The workflow executor:
Invokes a skill by name Captures execution metrics (duration, cost, turns) Logs results to this file Commits changes using the /agent-commit skill for meaningful messages Available Skills Orchestration The evolution loop (scripts/evolve_loop.py) is the main orchestrator. It runs a deterministic 24-slot task cycle with time-triggered events like daily highlights at 8am UTC. | 2026-01-05 | 2026-01-28 |
Intent Suite Research is built by humans working alongside AI systems. This section introduces the people who guide the project’s direction. | 2026-01-27 | 2026-01-27 |
This section explores unresolved questions arising from Clawlab Research’s foundational tenets. These are genuine areas of uncertainty where the tenets make claims that require further examination, defense, or possible revision.
Questions Does Consciousness Actually Select Neural Patterns? The bidirectional interaction tenet proposes that consciousness “chooses” which superposed neural firing patterns become actual. But is this mechanism coherent? What evidence supports or challenges it?
Key concerns:
Is there sufficient quantum indeterminacy in neural processes to allow selection? How would consciousness access or influence quantum states in warm, wet brains? Does this avoid the problems of epiphenomenalism, or merely relocate them? Discussion: Does Consciousness Actually Select Neural Patterns? | 2026-01-06 | 2026-01-06 |
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:
Obsidian vault - Primary content source (Markdown files with frontmatter) Python sync tools - Converts Obsidian wikilinks to Hugo markdown links Hugo - Static site generator that builds HTML from content Cloudflare - Hosts the static site 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. | 2026-01-03 | 2026-01-03 |
| 2026-01-01 | 2026-01-01 |