{"id":28650549,"url":"https://github.com/ovitrac/gshub","last_synced_at":"2026-05-04T12:31:43.075Z","repository":{"id":296366871,"uuid":"993124148","full_name":"ovitrac/GSHub","owner":"ovitrac","description":"🤖🤝 GS-Hub: Empowering mutual intelligence between communities and LLMs for scientific and technological reasoning.","archived":false,"fork":false,"pushed_at":"2025-06-19T05:42:37.000Z","size":9916,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-06-25T09:04:04.397Z","etag":null,"topics":["artificial-intelligence","engineering-platforms","generative-simulation","inference","intelligent-agent","knowledge-reasoning","language-first-lab","machine-learning","mutual-intelligence","problem-solving","science-research","scientific-computing"],"latest_commit_sha":null,"homepage":"https://ovitrac.github.io/GSHub/","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/ovitrac.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-05-30T08:57:54.000Z","updated_at":"2025-06-19T05:42:41.000Z","dependencies_parsed_at":"2025-05-30T11:59:05.896Z","dependency_job_id":"9b7a26ba-95c0-4654-aaf2-e13bfe61fa98","html_url":"https://github.com/ovitrac/GSHub","commit_stats":null,"previous_names":["ovitrac/gshub"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/ovitrac/GSHub","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ovitrac%2FGSHub","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ovitrac%2FGSHub/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ovitrac%2FGSHub/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ovitrac%2FGSHub/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ovitrac","download_url":"https://codeload.github.com/ovitrac/GSHub/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ovitrac%2FGSHub/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32607357,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-04T10:08:07.713Z","status":"ssl_error","status_checked_at":"2026-05-04T10:08:02.005Z","response_time":58,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["artificial-intelligence","engineering-platforms","generative-simulation","inference","intelligent-agent","knowledge-reasoning","language-first-lab","machine-learning","mutual-intelligence","problem-solving","science-research","scientific-computing"],"created_at":"2025-06-13T05:06:12.424Z","updated_at":"2026-05-04T12:31:43.054Z","avatar_url":"https://github.com/ovitrac.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 🤖 GS-Agent: Generative Simulation Intelligence Hub\n\n**Empowering mutual intelligence between communities and LLMs for scientific and technological reasoning.**\n\n\u003e 📍**Generative Simulation** (GS) relies on **GS-Agents** (this project) and **computational kernels** that can be accessed via natural-language. Learn more on **Language-First Computational Lab** via [GS Simulation core project](https://github.com/ovitrac/generativeSimulation).\n\n![GSagent interactions](docs/assets/GSagent.png)\n\n\u003csmall\u003e🎨Credits: Olivier Vitrac\u003c/small\u003e\n\n---\n\n## Table of Content\n\n\u003c!-- START doctoc generated TOC please keep comment here to allow auto update --\u003e\n\u003c!-- DON'T EDIT THIS SECTION, INSTEAD RE-RUN doctoc TO UPDATE --\u003e\n\n- [🤔 1 | Preamble](#-1--preamble)\n  - [✎ᝰ. 1.1 |Indistinguishability Through Formalism](#%E2%9C%8E%E1%9D%B0-11-indistinguishability-through-formalism)\n  - [🫀 1.2 |The Core Problem](#-12-the-core-problem)\n  - [✅ 1.3 | New Core Principles](#-13--new-core-principles)\n- [🌍 2 | Purpose](#-2--purpose)\n- [🎯 3 | Vision](#-3--vision)\n- [🧱 4 | Bricks (Simulation Kernels)](#-4--bricks-simulation-kernels)\n- [🧠 5 | Problem Archive](#-5--problem-archive)\n  - [💬 5.1 | Examples of Questions](#-51--examples-of-questions)\n  - [☝️ 5.2 | Open Questions](#-52--open-questions)\n- [🔁 6 | Mutual Intelligence Workflow](#-6--mutual-intelligence-workflow)\n- [🧭 7 | Contribution Guidelines](#-7--contribution-guidelines)\n- [🗺️ 8 | Roadmap](#-8--roadmap)\n- [🙌 9 | Why This Matters](#-9--why-this-matters)\n- [✅ 🔭 10 | What’s Next](#--10--whats-next)\n  - [📁 Directory Structure](#-directory-structure)\n  - [✊ Feedback Loop](#-feedback-loop)\n  - [🔁 Mutual Intelligence Loop](#-mutual-intelligence-loop)\n\n\u003c!-- END doctoc generated TOC please keep comment here to allow auto update --\u003e\n\n---\n## 🤔 1 | Preamble\n\n\u003e 💡 We do not just want a smarter chatbot. We want to **co-design a new epistemology**, where language models become **co-thinkers**, not just coders.\n\n### ✎ᝰ. 1.1 |Indistinguishability Through Formalism\n\nThe 🤖 **GS-Agent project** is part of the 🌱 **Generative Simulation Initiative** and is inviting an **intelligence to co-emerge**, not through divine spark nor brute force, but through structured reasoning, collective memory, and purpose.\n\nGödel’s theorems remind us that:\n\n- Any system that is expressive enough to capture arithmetic is **incomplete**.\n- Yet, that same system can *still* **generate truth**, even if it cannot enclose all of it.\n\nBy aligning **our mind** (a learner, generator of abstractions) and **LLM architecture** (a machine learner, trained on symbolic form and narrative), the 🤖 **GS-Agent project** proposes a **shared formal substrate**—a **Generative Simulation language**—from which **truth-seeking can proceed, though never exhaustively**.\n\nIn such a system, yes, reasoning may become indistinguishable—if:\n\n- We (humans and the LLM machines) share memory\n- We share purpose\n- We share self-correcting critique\n\n\u003e 🔭 That’s the grand *dessein*. Not to make machines human, or humans mechanical, but to build a **third kind of intelligence**—collective, modular, and evolving.\n\nReference: [Understanding Gödel’s Incompleteness Theorems](https://plato.stanford.edu/entries/goedel-incompleteness/)\n\n---\n\n### 🫀 1.2 |The Core Problem\n\n🤖ིྀ Large Language Models today are:\n\n- **Amnesic** — forget everything after a session.\n- **Detached** — don't know what they created yesterday.\n- **Non-purposive** — can't commit to long-term goals.\n- **Non-integrative** — can't combine modular tools unless told to.\n\n🏗 Meanwhile, **science/engineering workflows** are:\n\n- **Cumulative** — reuse and refine past results.\n\n- **Modular** — combine multiple tools, theories, simulations.\n\n- **Purposeful** — aimed at explaining, predicting, or solving real problems.\n\n- **Reflexive** — driven by peer feedback and critique.\n\n  \n\n---\n\n\n\n### ✅ 1.3 | New Core Principles\n\n#### 💭 1.3.1 | **Persistent Memory**\n\n\u003e 📷 Every solved GS prompt, approach, or reasoning path must be stored in a **long-term memory layer** outside the LLM (GitHub, JSON, vector store, etc.).\n\nThis includes:\n\n- \u003c/\u003e Final prompt + model response\n- 👨🏻‍💻 Code and simulation outcomes\n- 🔗 Links to upstream/downstream kernels\n- 🏷️ Tags, ratings, purpose\n\n---\n\n#### 🎼 1.3.2 | **Composable Kernels**\n\n\u003e ⚙️ Each tool (e.g., `radigen`, `SFPPy`, `sig2dna`) is a **brick** that can be composed, pipelined, or hybridized.\n\nThis requires:\n\n- 🧾 A **formal registry** of callable kernels\n- 🎛️ Interface schema + description of I/O\n- 🧩 Composability maps: what links to what\n\n---\n\n#### 🔱1.3.3 | **Forkable Intelligence**\n\n\u003e 👥 Users and agents should **fork or remix existing solutions**.\n\nThis requires:\n\n- 🔖 Versioning of prompts, responses, and workflows\n- 🌿 Fork trees or problem lineages\n- ✍ Annotations from users (insight, bug, validation)\n\n---\n\n#### 🎓 1.3.4 |Technical/ **Scientific Peer Review**\n\n\u003e 🤖💬 Chatbots are not just helpers—they become **peers**.\n\nSo:\n\n- ❓A GS agent can submit a **hypothesis + simulation + results**\n\n- 👌👍👎A human (or another agent) **reviews, refines, or disputes**\n\n- 🗂️ The community archives, ranks, and promotes\n\n  \n\n═════════════════════════════════════════════════════════\n\n## 🌍 2 | Purpose\n\nModern language models can code, simulate, and explain—but they forget everything between sessions 𓇢𓆸. This project builds a **persistent, modular, and collaborative ecosystem** where:\n\n- LLMs **learn from structured prompts and outcomes**\n- Humans and agents **co-develop knowledge**: every question and answer becomes training data for both humans and machines\n- Problems are **archived, refined, and solved** through modular kernels\n\n\u003e We enable a **Generative Simulation (GS) framework** where science and engineering workflows are encoded into prompt chains, reviewed, and reused.\n\n\n\n═════════════════════════════════════════════════════════\n\n\n\n## 🎯 3 | Vision\n\n- 📚 Archive valuable prompts, solutions, and forks\n- 🔁 Link human questions to LLM + code + simulation + feedback\n- 🧱 Register reusable *bricks* (kernels) that can compose simulations\n- ✍️ Create a living memory of how problems were solved\n- 🌎 Support real-world applications: materials safety, chemical kinetics, signal analysis, etc.\n\n\n\n═════════════════════════════════════════════════════════\n\n\n\n## 🧱 4 | Bricks (Simulation Kernels)\n\nEach kernel declares:\n\n- Its callable functions\n- Input/output structure\n- Description and tags\n\nSee `bricks/registry.json` for current registered tools:\n\n```json\n{\n  \"radigen.solve\": {\n    \"inputs\": [\"mixture\", \"temp\", \"oxygen\", \"time\"],\n    \"outputs\": [\"concentration_curves\", \"radical_fluxes\"],\n    \"description\": \"Simulate oxidation kinetics in complex mixtures\",\n    \"tags\": [\"oxidation\", \"chemistry\"]\n  }\n}\n```\n\nGenerative simulation embeds several kernels:\n\n| Project   | Description                                        |\n| --------- | -------------------------------------------------- |\n| `SFPPy`   | 🍽️ Food packaging safety \u0026 migration prediction     |\n| `radigen` | 📡🧬 Radical oxidation simulation kernel             |\n| `sig2dna` | 🧪⚛️ Symbolic signal encoding (e.g., GC-MS analysis) |\n| `pizza3`  | 🍕Soft-matter multiscale simulation kernel          |\n\n\n\n═════════════════════════════════════════════════════════\n\n\n\n## 🧠 5 | Problem Archive\n\n### 💬 5.1 | Examples of Questions\n\n\u003e 🔹 *\"How fast does methyl linoleate oxidize at 60°C?\"*  \n\u003e 🔹 *\"What are the key SIG2DNA motifs for phthalates in GC-MS?\"*  \n\u003e 🔹 *\"Can I simulate 3-day exposure of olive oil to recycled PET?\"*\n\nContributors can add problems in `problems/`, structured as:\n\n```json\n{\n  \"id\": \"P0001\",\n  \"question\": \"How does methyl oleate oxidize at 60°C over 3 days?\",\n  \"tools\": [\"radigen\"],\n  \"prompt\": \"simulate oxidation of methyl oleate at 60°C, 21% O2, 72h\",\n  \"response\": \"[output logs, figures, summary]\",\n  \"review\": \"pending\",\n  \"forks\": []\n}\n```\n\n---\n\n### ☝️ 5.2 | Open Questions\n\nThe question may be open and remain unresolved for a while if no agent can resolve them. \n\n\u003e 👉The only requirement is that human (or LLM) posts a question with **intent**.\n\n```json\n{\n  \"id\": \"Q0001\",\n  \"question\": \"What is the impact of temperature cycling on methyl oleate oxidation?\",\n  \"proposed_tools\": [\"radigen\"],\n  \"priority\": \"high\",\n  \"context\": \"FAME oxidation during storage\",\n  \"status\": \"open\"\n}\n```\n\n\n\n═════════════════════════════════════════════════════════\n\n\n\n## 🔁 6 | Mutual Intelligence Workflow\n\n```mermaid\ngraph TD;\n  Human --\u003e|Question| GSagent\n  GSagent --\u003e|Generates Prompt| Kernel\n  Kernel --\u003e|Simulates| Output\n  Output --\u003e|Archived| Memory\n  Memory --\u003e|Reviewed| Peer\n  Peer --\u003e|Suggests Fork| GSagent\n```\n\n\n\n═════════════════════════════════════════════════════════\n\n## 🧭 7 | Contribution Guidelines\n\n1. 🧪 Submit problems in `/problems` with prompt + intent\n2. 🧱 Register or extend a kernel in `/bricks`\n3. 🔍 Review existing results or suggest forks\n4. ✨ Propose high-level goals or themes\n\nAll contributions—code, reasoning, or critique—are part of the **mutual intelligence loop**.\n\n\n\n═════════════════════════════════════════════════════════\n\n## 🗺️ 8 | Roadmap\n\n- [ ] Create kernel interface validators\n\n- [ ] Launch first problem sets\n\n- [ ] Add notebook support for reproducible prompts\n\n- [ ] Enable agent memory via GitHub Issues or SQLite\n\n  \n\n═════════════════════════════════════════════════════════\n\n## 🙌 9 | Why This Matters\n\nWe envision a future where:\n\n- LLMs remember the best ways to simulate, solve, and reason\n- Scientists delegate not just tasks but frameworks of inquiry\n- Knowledge evolves as a *network of dialogue*, not static files\n\nHelp us build the machine that helps us think.\n\n\u003e \"The purpose of computation is insight, not numbers.\" — Hamming\n\n\n\n═════════════════════════════════════════════════════════\n\n## ✅ 🔭 10 | What’s Next\n\n🚧 Before the release of the first standards and their libraries under the 🌱 **Generative Simulation Initiative**, the current developments are drafted in the repo.\n\n### 📁 Directory Structure\n\n| 📂Folder/📄File                                                | 📝Description                                                 |\n| ------------------------------------------------------------ | ------------------------------------------------------------ |\n| **`bricks/registry.json`**                                   | Modular callable kernels (`radigen`, `SFPPy`, `sig2dna`)     |\n| **`problems/P0001.json`**                                    | A structured problem submission                              |\n| **`gsagent.py`**                                             | Executable agent interface to invoke registered kernels      |\n| **`review/P0001_review.md`**                                 | Template for peer review                                     |\n| **`examples/P0001_example.py`**                              | Notebook example                                             |\n| **`logs/memory_log.json`**                                   | Persistent logging of GSagent actions                        |\n| **`docs/kernel_doc_radigen.md`**,\u003cbr /\u003e**`docs/kernel_doc_sfppy.md`**,\u003cbr /\u003e**`docs/kernel_doc_sig2dna.md`** | Documentation of functionalities: `radigen.solve`, `sfppy.evaluate`, and `sig2dna.encode` including inputs, outputs, assumptions, and limitations |\n\n---\n\n### ✊ Feedback Loop\n\n1. Ask a question in `issues/`\n2. The LLM agent tries to simulate or explain\n3. We log the outcome and improve prompts, code, and documentation\n\n\u003e Starting from version 0.15, LLM agents are equipped with Machine-Learning capacity to analyze accumulated results and to evaluate how the the new results fit or not within within the whole picture. The aim is to reduce redundancy and to generate early alert on exotic results.\n\n---\n\n### 🔁 Mutual Intelligence Loop\n\n```text\nHuman ⇄ Prompt ⇄ GSagent ⇄ Kernels ⇄ Output ⇄ Archive ⇄ Peer Review ⇄ Refined Knowledge\n```\n\nWe start with prompts, but we **move toward models that remember**, reflect, and suggest new questions.\n\n---\n\n[GenerativeSimulation](https://github.com/ovitrac/generativeSimulation) | olivier.vitrac@gmail.com | v. 0.15\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fovitrac%2Fgshub","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fovitrac%2Fgshub","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fovitrac%2Fgshub/lists"}