A structured system for conducting independent physics-based technology due diligence and risk assessment on hard-tech startup investments.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install v1-0-0
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 v1-0-0 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 Deeptechnic Deep-Tech Due Diligence Framework?
Deeptechnic is a comprehensive deep-tech due diligence framework designed to guide AI agents and investment teams through structured, independent technical validation of hard-tech startups. Instead of relying solely on pitch decks or business plans (BPs), Deeptechnic implements a rigorous evaluation methodology across 10 specific quality dimensions. It empowers teams to verify physical formulas, map out engineering migration gaps, and audit supply chain vulnerabilities before committing capital.\n\nBy leveraging this framework among other Openclaw Skills, developers and investment analysts can coordinate multiple dedicated sub-skills—such as team capability assessment, competitor analysis, and automated feedback collection. This ensures a thorough, bias-free, and highly quantitative analytical process for checking complex technical claims.
Deeptechnic Deep-Tech Due Diligence Framework Use Cases
- Evaluating Deep-Tech Claims: Perform physics-based validation and calculations on a startup's core technology claims to prevent investing in impossible science.\n- Pre-Investment Tech Risk Assessment: Conduct first-pass technical assessments before starting expert interviews, mapping out high-priority gaps.\n- Team Capability and Gap Mapping: Analyze team members individually to match their credentials against real-world execution requirements.\n- Supply Chain and Geopolitical Auditing: Identify supplier-level vulnerabilities and verify the availability of specific components instead of generic domestic alternatives.\n- Automated Due Diligence Report Generation: Automatically compile multi-round evaluation findings into structured executive reports.
How Deeptechnic Deep-Tech Due Diligence Framework Works
- Round 1 (First Pass): Collect public patents, papers, and news to validate core technical claims with math and physics equations, identify initial team gaps, and draft priority questions.\n2. Round 2 (Team Feedback): Refresh the evaluation model with direct team answers to construct expert interview outlines and precise supply chain queries.\n3. Round 3 (External Input): Gather and cross-validate multiple external sources, then record findings from expert interviews.\n4. Round 4 (Final Synthesis): Compile all findings into a structured report following the predefined structure and name-spacing.\n5. Quality Gate Validation: Execute the 10-point quality gate to score the output against strict metric-based benchmarks (0.0 to 1.0) before concluding the workflow.
Deeptechnic Deep-Tech Due Diligence Framework Setup
To begin using Deeptechnic within Openclaw Skills, initialize the framework by loading the skills index and sequential sub-skills:\n\nbash\n# Step 1: Discover available sub-skills in the index\nload references/skills-index.md\n\n# Step 2: Initialize Round 1 analysis and anti-rationalization validation\nload references/skills/deep-tech-dd-analysis\nload references/skills/anti-rationalization\n\n# Step 3: Load Round 2 skills for team and supply chain audits\nload references/skills/team-assessment\nload references/skills/supply-chain-audit\n\n# Step 4: Load specialized agent skills for competitive analysis\nload references/agents/tech-reviewer\nload references/agents/competition-analyst\n\n# Step 5: Activate quality gate checks before final report generation\nload references/skills/quality-gate\n
Deeptechnic Deep-Tech Due Diligence Framework Data Schema & Taxonomy
Scoring Metric (10 Quality Dimensions)\n\nEvery due diligence report gets systematically evaluated across these quantitative axes:\n\n| Dimension | Weight | Target Focus |\n| :--- | :--- | :--- |\n| claim_validation | 0.20 | Physics-based proof and formulas verification |\n| engineering_migration_gap | 0.15 | Lab-to-product engineering timeline risks |\n| per_person_assessment | 0.15 | Person-by-person capabilities matching |\n| independent_competitor | 0.10 | Non-disclosed competitor discovery |\n| supplier_level_analysis | 0.10 | Specific tier-1 and tier-2 vendor verification |\n| independent_analysis | 0.10 | Ratio of independent facts vs BP-sourced facts |\n| team_gap_precision | 0.05 | Identification of structural hiring needs |\n| quantitative_score | 0.05 | Decimal-precision rating system |\n| priority_levels | 0.05 | Sorting open questions by P0/P1/P2 priorities |\n| overview_page | 0.05 | Front-page core matrix summary |\n\n### Report File Structure\n\nGenerated reports are saved using the format Deeptechnic_<ProjectEngName>_<Phase>_<Date>.docx containing:\n- Executive overview and claim judgment table.\n- Independent technical feasibility analysis.\n- Granular team and supplier maps.\n- Competitor risk matrix.
Deeptechnic Deep-Tech Due Diligence Framework Advanced Features
- Anti-Rationalization Engine: Built-in guardrails to actively combat confirmation bias, ensuring the AI agent does not gloss over unverified founder claims.\n- Multi-Agent Reference Bundles: Orchestrate highly specialized agents (e.g., Tech Reviewer and Competition Analyst) for specialized validation tasks.\n- Strict Quality Gates: Programmatic validation checklist requiring perfect compliance across all 10 core rules before finalized output delivery.\n- Feedback Collection Loop: Saves post-evaluation user feedback directly to a localized training dataset (
SkillOpt) for automated agent self-improvement.
SKILL.md
Loading
Related Openclaw Skills
Deckly Deck Redesign
An interactive AI coding agent skill to automatically redesign, beautify, and polish PowerPoint (.pptx) and PDF slides using the Deckly API.
Literature Work Page Generator
An AI coding agent skill designed to automatically generate feature-rich, standalone HTML presentation and analysis pages for literary and creative works.
FastAPI Web Framework
A high-performance Python Web API framework skill that enables AI agents to design, generate, and deploy modern, production-ready RESTful APIs with native async support.
PDF Processing Skill
A versatile skill for reading, writing, merging, splitting, OCR-ing, and extracting tables from PDF files using Python and command-line tools.
Multi-Platform Social Media Data Scraper
An elite multi-platform scraper to capture social media metrics and automatically sync them to Feishu Bitables.
Domain Payload Generator
Domain Payload Generator is a specialized meta-skill designed to analyze new technical fields and independently manufacture fully UTOS-compatible domain payload skills from scratch.
Featured*
