An advanced analytical skill that strips claims down to irreducible truths and rebuilds them without unverified assumptions or flawed analogies.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install deciqai-first-principles
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install deciqai-first-principles using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is First Principles?
The First Principles skill is a structured reasoning tool designed for AI agents to dismantle conventional assumptions, analogies, and unverified authority. Rather than reasoning by analogy—which risks inheriting systemic errors and expensive, hard-to-reverse biases—this skill decomposes a problem down to its core, irreducible statements such as physical laws, mathematical identities, explicit definitions, or citable empirical facts.
Once the problem is stripped down to foundational bedrock, the skill guides the agent to rebuild the solution from scratch. It functions as one of three composable core motions in cognitive engineering: decomposing downward to bedrock, while complementary workflows select the simplest path or project consequences forward through time.
First Principles Use Cases
- Pressuring-testing expensive, high-risk, or hard-to-reverse decisions involving architecture, business models, or pricing.
- Deconstructing justifications that rely heavily on "best practices," industry standards, or standard authority.
- Structuring complex knowledge artifacts from irreducible elements before initiating long-form technical writing.
- Triggering explicit deeper analysis when a user asks to evaluate a concept from scratch or tear a conventional answer apart.
How First Principles Works
- State the claim: Formulate the primary decision, belief, or assertion into a single, concise sentence.
- Surface assumptions: Thoroughly and uncharitably list every hidden or explicit assumption underlying the claim.
- Tag components: Categorize every single assumption as either BEDROCK (irreducible or citable) or INHERITED (convention, analogy, or unverified authority).
- Interrogate items: Systematically audit all INHERITED tags to determine if they are necessarily true, leading to a DEMOLISHED status or promotion to BEDROCK.
- Terminate regress: Stop the deconstruction precisely at indemonstrable or citable facts to avoid analysis paralysis.
- Reconstruct solution: Rebuild the assertion, architecture, or strategy using exclusively surviving bedrock elements.
- Diff against convention: Output a clear comparative analysis mapping what changed and detailing exactly why.
First Principles Setup
To deploy this skill within your environment, incorporate the configuration directly into your agent's system prompt or runtime skills directory.
# Clone your Openclaw Skills repository context
mkdir -p skills/cognitive
cp first-principles.md skills/cognitive/
# Validate the skill pathing dependencies
ls skills/cognitive/bedrock-packs/
Configure the delivery context by setting the operational mode parameter to either engine for direct execution or coach for interactive, step-by-step user instruction.
First Principles Data Schema & Taxonomy
The skill processes inputs dynamically and outputs a structured execution artifact following a rigid taxonomy.
Output Template Structure
| Field | Description | Type |
|---|---|---|
Claim |
A single-sentence encapsulation of the core premise | String |
Assumptions |
Fully exhaustively listed factors tagged as BEDROCK or INHERITED |
Key-Value List |
Bedrock |
The verified irreducible or cited facts used as foundations | List |
Reconstruction |
The rebuilt thesis or architecture derived from valid inputs | String |
What Changes |
An explicit delta report contrasting the new output against convention | Markdown String |
Confidence |
Open questions, risks, and boundary metrics for execution | List |
Extension Domain Packs
Domain-specific logic is isolated into modular bedrock packs:
business-unit-economics.md: Foundation models for pitches, revenue structures, and commercial validation.knowledge-domain-decomposition.md: Foundation rules for organizing systemic knowledge bases before code generation.
First Principles Advanced Features
- Adaptive Front Door: Intelligently switches delivery behaviors between Engine Mode for instant automated teardowns and Coach Mode for guided, multi-turn human-in-the-loop training.
- Hard-Stop Halting: Utilizes structured stop gates to prevent the AI agent from expanding assertions before the human validator explicitly verifies the prior analytical tier.
- Anti-Analogy Guardrails: Active runtime validation blocks common rationalizations, preventing the smuggling of unverified market numbers, emotional sunk costs, or hidden analogies back into the final architectural reconstruction.
SKILL.md
Loading
Related Openclaw Skills
Bayesian Reasoning
A systematic decision-making framework that applies Bayes' theorem to prevent cognitive biases, base-rate neglect, and prosecutor's fallacies during complex evidence interpretation.
OPC Zhinao
OPC Zhinao is a full-stack startup diagnosis agent designed for solo entrepreneurs, guiding them through a quantified five-stage validation and growth framework.
Lamb (The Gospel According to Biff)
An AI coding agent skill that maps Christopher Moore's Lamb to user queries about deep friendship, coping with humor, accepting tragedy, and finding one's purpose.
Lamb: The Gospel According to Biff - AI Companion Skill
An AI agent skill based on Christopher Moore's Lamb that helps users navigate complex themes of friendship, faith, humor, and purpose through the lens of Joshua and Biff.
Wintergirls - Eating Disorder Support & Recovery Guide
An AI companion skill that leverages the deep narrative of Laurie Halse Anderson's Wintergirls to guide users through understanding eating disorders, managing grief, and navigating the difficult path to mental health recovery.
DOCX
An advanced operational skill for creating, editing, and extracting rich data from Microsoft Word documents using automated tooling and XML modification.
Featured*
