Back to search
Agent-Native-Research-Artifact
SkillGitHub
Research Artifact Protocol for Rigorous and Trustworthy AI Scientists
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Checked today
Badge for your README
Markdown
[](/listings/ara-labs--agent-native-research-artifact)HTML
<a href="/listings/ara-labs--agent-native-research-artifact"><img src="/api/listings/ara-labs--agent-native-research-artifact/badge.svg" alt="findsafeskills safety rating" /></a>Install
i
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
README.md · README · 188 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- # 🔬 Agent-Native Research Lab
- ### The Essential Toolkit to Supercharge Your Autoresearch
- [](#quickstart)
- [](LICENSE)
- [](skills/)
- [](https://arxiv.org/abs/2604.24658)
- [](docs/poster.pdf)
- [](https://github.com/ARA-Labs/ARA-Demo)
- > **Autoresearch is blindingly fast, but fundamentally unobservable.**
- > When AI agents run experiments, they leave behind a graveyard of overwritten code, scattered logs, and undocumented dead ends. ARA is the essential toolkit that forces your AI scientists to work in a **structured, verifiable, and traceable** format—so you can actually trust the science they produce without reverse-engineering thousands of lines of terminal output.
- <p align="center">
- <img src="docs/ara-skills-demo.gif" alt="ARA Skills Demo" width="100%"/>
- </p>
- ---
- <a id="quickstart"></a>
- ## ⚡ Start here — the only setup step
- **Paste this one line into your coding agent, from your project root:**
- ```
- Read https://raw.githubusercontent.com/ARA-Labs/Agent-Native-Research-Artifact/main/wire-ara.md and follow its instructions.
- ```
- The agent fetches the full wiring prompt, installs the seven skills, and writes a task-routing map into its own context file (`CLAUDE.md`, `AGENTS.md`, `.cursorrules`, or `GEMINI.md`) — so every future session knows which skill handles which situation to assist your ai-native research. The prompt lives at [`wire-ara.md`](wire-ara.md).
- Using ARA as your **conference submission format**? Follow the short guide: **[Using ARA as a Conference Submission Format](docs/conference-guide.md)** — set up, record as you research, publish as one link.
- ---
- ## Core Design Principles
- Instead of leading with layers, the bundle maps directly to how it solves the bottleneck through three core design principles:
- <p align="center">
- <img src="docs/figures/fig_three_principles.png" alt="The three core design principles: Guardrailing & Verification, Crystallizing Insights, and Total Observability" width="100%"/>
- </p>
- <details>
- <summary><b>🛡️ Guardrailing & Verification</b> — every claim wired to ground-truth execution</summary>
- <br/>
- AI agents require precise constraint boundaries to prevent hallucinated conclusions. The system acts as a strict **epistemic anchor**, automatically applying formal verification principles to ensure every scientific claim is directly wired to ground-truth execution and falsifiable results.
- </details>
- <details>
- <summary><b>🧠 Crystallizing Insights</b> — messy trajectories become structured knowledge</summary>
- <br/>
- Research is rarely a straight line; it is a messy graph of pivots and dead ends. The system forces AI scientists to systematically document their trajectory, crystallizing fleeting, unstructured logs into highly structured, reliable research knowledge that builds compounding value over time.
- </details>
- <details>
- <summary><b>👁️ Total Observability</b> — high-level oversight without reading terminal output</summary>
- <br/>
- Supervising AI scientists shouldn't require reading endless terminal outputs. The system translates complex agent behaviors and exploration graphs into a clean, minimalist interface. It lets human researchers maintain high-level oversight, seamlessly stepping in to course-correct or guide the AI's behavior with zero friction.
- </details>
- ## 🛠️ The Toolkit: Seven Core Skills
- To operationalize these design principles, ARA provides a suite of seven specialized agent skills. The one-liner above installs all seven automatically; to install manually or pick individual skills:
- ```bash
- npx @ara-commons/ara-skills
- ```
- Auto-detects Claude Code, Cursor, Gemini CLI, OpenCode, Codex, and Hermes, then prompts for skills, agents, and install scope (global vs. local). Full CLI reference: [`packages/ara-skills/`](packages/ara-skills/).
- Then reach for a skill by what you need:
- | If you want to… | Skill | Invoke |
- |---|---|---|
- | **Capture** research faithfully as you work — decisions, ablations, dead ends, configs | **research-manager** | `/research-manager` (or wire it to run automatically) |
- | **Compile** an existing paper, repo, or notes into a structured ARA | **compiler** | `/compiler <path>` |
- | **Verify** an artifact's epistemic rigor before you trust, publish, or submit it | **rigor-reviewer** | `/rigor-reviewer <dir>` |
- | **Observe** the full research trajectory in an interactive process map — a portable narrated HTML export, or a live local viewer with validation/linting via the official [`ara` CLI](https://github.com/ARA-Labs/ara-cli) | **research-visualizer** | `/research-visualizer <ara-dir>` (add `--serve` or `--check` for the live/lint path) |
- | **Ask** an ARA anything — grounded, falsifiable answers to "what should I try next / why did this work / what if I change X" ([demo](https://www.agenticresearch.sh/blog/research-world-model)) | **research-foresight** | `/research-foresight <ara-dir> "<question>"` |
- | **Share** a file or folder with someone else's agent — one URL they paste, no repo, no zip | **context-drop** | `/context-drop <path>` |
- | **Steer** any open-ended investigation the way a fuzzer steers inputs — predict before every action, track untried leads and unexplained results, alarm on going-in-circles, gate every conclusion behind a self-refutation attempt ([repo](https://github.com/ARA-Labs/research-fuzzer)) | **research-fuzzer** | `/research-fuzzer` (fire at investigation start, after each batch of results, before any conclusion) |
- See each skill's `SKILL.md` for the full specification:
- [research-manager](skills/research-manager/SKILL.md) ·
- [compiler](skills/compiler/SKILL.md) ·
- [rigor-reviewer](skills/rigor-reviewer/SKILL.md) ·
- [research-visualizer](skills/research-visualizer/SKILL.md) ·
- [research-foresight](skills/research-foresight/SKILL.md) ·
- [context-drop](skills/context-drop/SKILL.md) ·
- [research-fuzzer](skills/research-fuzzer/SKILL.md)
- ---
- ## Under the hood — the artifact anatomy
- The four pillars all read and write one structure. An ARA organizes research into four interlocking layers:
- ```
- example_artifact/
- PAPER.md # Root manifest + layer index (~200 tokens)
- logic/ # Cognitive layer — What & Why
- claims.md # Falsifiable assertions with proof refs
- experiments.md # Declarative experiment plans
- solution/
- architecture.md # System design + component graph
- algorithm.md # Math + pseudocode
- constraints.md # Boundary conditions
- related_work.md # Typed dependency graph
- src/ # Physical layer — How
- configs/ # Hyperparameters with rationale
- environment.md # Dependencies, hardware, seeds
- trace/ # Exploration graph — Journey
- exploration_tree.yaml # Research DAG with typed nodes + dead ends
- evidence/ # Raw proof
- tables/ # Exact result tables
- figures/ # Extracted data points
- ```
- <p align="center">
- <img src="docs/figures/fig_cross_layer_v2_attempt1.png" alt="Cross-layer bindings" width="90%"/>
- </p>
- *Cross-layer forensic bindings thread claims in `/logic` to code in `/src` and evidence in `/evidence`. Dead-end nodes (×) in the exploration graph preserve failure modes so no agent re-walks them.*
- **Key structural principles**
- - **Progressive disclosure** — `PAPER.md` (~200 tokens) tells an agent whether the artifact is relevant; deeper files load on demand.
- - **Cross-layer binding** — claims reference experiments, experiments reference evidence, heuristics reference code. Everything resolves.
- - **Dead ends preserved** — failed approaches and rejected alternatives are first-class nodes in the exploration graph, not noise to drop.
- - **Provenance tracking** — every entry is tagged (`user`, `ai-suggested`, `ai-executed`, `user-revised`), distinguishing human-confirmed facts from AI inferences.
- ---
- ## Why it works
- The supervision gap is not hand-waving — it shows up as measurable cost. Across benchmarks, an ARA beats a strong PDF + repo baseline on the three things agents do with research (understand, reproduce, extend), most dramatically on recovering the *failure* knowledge a narrative drops. For the full argument — the two structural taxes, the benchmark results, and the case for agent-native research — read the writeup:
- **→ [The Last Human-Written Paper: Agent-Native Research Artifacts](https://amberljc.github.io/blog/2026-04-24-the-last-human-written-paper.html)**
- This paper practices what it proposes — its own ARA lives at [`examples/the-ara-of-ara`](examples/the-ara-of-ara).
- ---
- ## Compatibility
- These skills follow the [Agent Skills open standard](https://agentskills.io/specification) and work with:
- - [Claude Code](https://claude.ai/code) (Anthropic)
- - [Codex CLI](https://github.com/openai/codex) (OpenAI)
- - [GitHub Copilot](https://github.com/features/copilot)
- - [Cursor](https://cursor.com)
- - Any agent supporting the Agent Skills specification
- ---
- ## Citation
- If you use ARA in your research, please cite:
- ```bibtex
- @misc{liu2026humanwrittenpaperagentnativeresearch,
- title={The Last Human-Written Paper: Agent-Native Research Artifacts},
- author={Jiachen Liu and Jiaxin Pei and Jintao Huang and Chenglei Si and Ao Qu and Xiangru Tang and Runyu Lu and Lichang Chen and Xiaoyan Bai and Haizhong Zheng and Carl Chen and Zhiyang Chen and Haojie Ye and Yujuan Fu and Zexue He and Zijian Jin and Zhenyu Zhang and Shangquan Sun and Maestro Harmon and John Dianzhuo Wang and Jianqiao Zeng and Jiachen Sun and Mingyuan Wu and Baoyu Zhou and Chenyu You and Shijian Lu and Yiming Qiu and Fan Lai and Yuan Yuan and Yao Li and Junyuan Hong and Ruihao Zhu and Beidi Chen and Alex Pentland and Ang Chen and Mosharaf Chowdhury and Zechen Zhang},
- year={2026},
- eprint={2604.24658},
- archivePrefix={arXiv},
- primaryClass={cs.LG},
- url={https://arxiv.org/abs/2604.24658},
- }
- ```
- ---
- ## Contributing
- See [CONTRIBUTING.md](CONTRIBUTING.md) for how to add or improve skills.
- ## License
- [MIT](LICENSE)
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.