A high-rigor research assistant designed to deconstruct technical papers into durable, evidence-based markdown notes and implementation memos.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install paper-deep-reader
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 paper-deep-reader 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 Paper Deep Reader?
Paper Deep Reader is a sophisticated framework within the Openclaw Skills ecosystem dedicated to serious academic and technical reading. Unlike basic summarizers that simply rewrite abstracts, this skill treats paper reading as a process of reconstruction and judgment. It is specifically engineered for graduate-level analysis of complex documents such as arXiv preprints, journal articles, and technical reports.
The skill excels at capturing the technical spine of a paper—including equations, notations, derivations, and experimental benchmarks—across fields like machine learning, quantitative finance, systems engineering, and physics. By forcing a structured routing and evidence-checking protocol, it ensures that every claim is traced back to specific data, figures, or appendix evidence, providing a level of depth that goes far beyond surface-level AI summaries.
Paper Deep Reader Use Cases
- Deep reading of complex machine learning or physics papers requiring equation reconstruction.
- Generating technical implementation memos for algorithms or system designs.
- Critiquing empirical identification strategies in economics or social sciences.
- Creating notation tables and claim-evidence matrices for high-stakes research projects.
- Reconstructing the intellectual structure of a paper to identify hidden limitations or failure risks.
How Paper Deep Reader Works
- Paper Mapping: Before drafting, the skill builds an internal map identifying the research question, problem setting, and the primary 'intellectual load' of the work.
- Strategic Routing: Using Openclaw Skills routing rules, it selects a primary adapter (e.g., Theory, Systems, or Empirical Econ) and up to three evidence checklist packs.
- Multi-Pass Reading: The skill executes a 4-pass reading strategy: Framing, Technical Core, Evidence, and Limits/Context.
- Structured Artifact Generation: It generates mandatory internal structures including notation tables, claim-evidence matrices, and limitation ledgers.
- Note Synthesis: Finally, it renders a rigorous markdown note that distinguishes between the authors' claims and objective technical evaluation.
Paper Deep Reader Setup
To initialize the environment for this skill, ensure the core reference files and scripts are accessible in your base directory:
# Standard setup for Paper Deep Reader within Openclaw Skills
# 1. Verify directory structure
ls {baseDir}/references/adapters/
ls {baseDir}/scripts/
# 2. Recommended sequence for generating a deep note
python3 scripts/scaffold_note.py --paper path/to/paper.pdf
python3 scripts/build_paper_map.py
python3 scripts/build_notation_table.py
python3 scripts/render_final_note.py
Paper Deep Reader Data Schema & Taxonomy
The skill organizes research data into a strictly structured markdown format based on the following taxonomy:
| Component | Description |
|---|---|
| Paper Map | Problem, setting, main move, and primary failure risks. |
| Route Record | Selected adapters, evidence packs, and route confidence. |
| Notation Table | Symbol, meaning, domain, and units of technical objects. |
| Claim-Evidence Matrix | Direct mapping of author claims to specific supporting evidence. |
| Limitation Ledger | Explicitly stated vs. inferred reader limitations. |
Paper Deep Reader Advanced Features
- Multi-adapter routing for interdisciplinary papers where contributions span multiple domains.
- Integration with evidence checklist packs for rigorous validation of proof rigor and experimental fairness.
- Domain Overlays for specialized fields with recurring technical traps or overclaims.
- Automated script-based drafting aids for notation extraction and limitation tracking.
- Appendix-aware analysis to ensure claims are supported by supplemental evidence often missed by basic LLMs.
SKILL.md
Loading
Related Openclaw Skills
Piston Reference Skill
A local, high-speed reference tool for developers to access Piston concepts, implementation patterns, and best practices directly within their AI agent workflow.
Code → PRD
A powerful reverse-engineering tool that transforms frontend, backend, or fullstack codebases into comprehensive, business-readable Product Requirements Documents.
Markdown Converter
A versatile document processing utility that transforms complex file formats into clean, structured markdown optimized for LLM analysis.
China Express Tracker
A precision logistics tracking skill that uses browser automation to retrieve real-time Chinese package statuses directly from Kuaidi100.
Servo
A local, zero-dependency reference documentation tool for developers providing instant access to technical patterns and guides.
Agent Pulse
A standardized protocol for real-time AI agent status checks, providing instant visibility into load, interruptibility, and task readiness.
Featured*
