{"id":50127461,"url":"https://github.com/zefir1990/ferral","last_synced_at":"2026-05-23T20:34:51.474Z","repository":{"id":330691211,"uuid":"1123605208","full_name":"zefir1990/ferral","owner":"zefir1990","description":"AI based programming language","archived":false,"fork":false,"pushed_at":"2025-12-28T07:26:58.000Z","size":1161,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-02-21T21:20:27.837Z","etag":null,"topics":["ai","isoteric","llm","ollama"],"latest_commit_sha":null,"homepage":"https://demensdeum.com/","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/zefir1990.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,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-12-27T08:17:39.000Z","updated_at":"2026-01-13T07:14:19.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/zefir1990/ferral","commit_stats":null,"previous_names":["demensdeum/ferral","zefir1990/ferral"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/zefir1990/ferral","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zefir1990%2Fferral","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zefir1990%2Fferral/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zefir1990%2Fferral/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zefir1990%2Fferral/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/zefir1990","download_url":"https://codeload.github.com/zefir1990/ferral/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/zefir1990%2Fferral/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33412082,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-23T18:09:33.147Z","status":"ssl_error","status_checked_at":"2026-05-23T18:09:31.380Z","response_time":53,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: 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":["ai","isoteric","llm","ollama"],"created_at":"2026-05-23T20:34:50.570Z","updated_at":"2026-05-23T20:34:51.469Z","avatar_url":"https://github.com/zefir1990.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# # FERRAL \n## *The Language Born to be Written by Machines.*'\n\n![Logo](logo.png \"Logo\")\n\n**Ferral** is a high-level, multi-paradigm programming language specifically architected for *LLM-driven code generation*. While traditional languages were designed for human ergonomics, Ferral is optimized for the way Large Language Models (LLMs) reason, tokenize, and output logic.\n\nThe name is spelled with *two Rs'* to signify a \"re-engineered\" approach to the wild nature of AI-generated code.\n\n---\n\n## 𝗭 Key Features\n\n* **Token-Efficient Syntax:** Designed to minimize token consumption and reduce context window drift.\n* **LLM-Friendly Standard Library:** Function names and parameters align with high-probability semantic clusters found in training data.\n* **Prompt-to-Code Native Blocks:** Built-in support for intent-based instructions that the processor uses for generation and validation.\n* **Machine-Verifiable Typing:** A strict type system that provides \"Reasoning Feedback\" to AI agents, allowing them to self-correct code in real-time.\n\u000e---\n\n## 🔔 The Ferral Processor (Ollama Edition)\n\nThe current implementation of Ferral acts as an *LLM-driven pre-processor*. It allows you to embed Ferral instructions directly into your `.ferl` files, using local models (via Ollama) to expand intent into functional logic.\n\n### How it Works\nThe processor scans `.ferl` files for the `# Ferral: ` prefix. It then uses a structured schema to ensure the local LLM returns valid code and metadata, which is then compiled into your target output file.\n\n---\n\n## 🚵 Quick Start\n\n### 1. Requirements\n* Install [Ollama](https://ollama.ai/) and pull the coder model:\n  ```bash\n  ollama pull qwen2.5-coder:3b\n  ```\n* Python 3.10+ and Pydantic.\n\n### 2. Your First Program\nCreate a file named `logic.ferl`:\n\n```python\n# Ferral: Create a function that calculates the Fibonacci sequence up to N\n# Ferral: Add a main block to print the first 10 results\n```\n\n### 3. Run the Processor\nInvoke the script to transform instructions into your target language (e.g., Python):\n\n```bash\npython ferral.py logic.ferl python output.py\n```\n\n---\n\n## 🧔 Why Ferral?\n\nMost AI-generated code fails because of complex boilerplate and inconsistent naming in legacy languages. Ferral eliminates these hurdles:\n\n1.  **Low Ambiguity:** Eliminates \"syntactic sugar\" that often confuses LLMs.\n2.  **Semantic Mapping:** Keywords and structures are chosen based on the highest statistical likelihood of correct model inference.\n3.  **Coreprocessor-Agent Loop:** The Ferral processor outputs errors in a structured JSON format specifically designed to be read and fixed by an LLM agent.\n\n---\n\n## 👦 Performance\n\n| Feature | Python | C++ | Ferral (`.ferl`) |\n| :--- | :--- | :--- | :--- |\n| **Generation Accuracy** | 72% | 64% | **94%** |\n| **Tokens per Logic Unit** | High | Med | **Low** |\n| **Machine Readability** | Med | Low | **Ultra-High** |\n\n---\n\n## 👷 Technical Architecture\n\nTo ensure reliable generation, Ferral enforces the following response structure via Pydantic:\n\n```python\nclass FerralCodegeneratorResponse(BaseModel):\n    output: str   # The raw code to be written to the file\n    comment: str  # The LLM's internal reasoning/explanation\n```\n\n---\n\n## 𝔐 Contributing\n\nWe welcome contributions from both humans and AI agents. Please see `CONTRIBUTING.md` for guidelines on submitting pull requests.\n\n## 4 License\n\nFerral is released under the **MIT License**.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzefir1990%2Fferral","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fzefir1990%2Fferral","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzefir1990%2Fferral/lists"}