{"id":49670070,"url":"https://github.com/vedantjadhav701/developer-code-intelligence-agent","last_synced_at":"2026-05-06T22:02:01.494Z","repository":{"id":355922552,"uuid":"1230212474","full_name":"VedantJadhav701/Developer-Code-Intelligence-Agent","owner":"VedantJadhav701","description":"🧠 Local AI coding agent that autonomously finds bugs, writes fixes, self-reviews, and validates with tests. Powered by Ollama. 100% offline. Zero API costs.","archived":false,"fork":false,"pushed_at":"2026-05-05T21:15:41.000Z","size":52,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-05T22:27:09.069Z","etag":null,"topics":["ai-agent","autonomous-agent","code-intelligence","coding-agent","developer-tools","local-llm","offline-ai","ollama","python","react-loop","self-review"],"latest_commit_sha":null,"homepage":"","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/VedantJadhav701.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":"CONTRIBUTING.md","funding":".github/FUNDING.yml","license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","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},"funding":{"github":["VedantJadhav701"]}},"created_at":"2026-05-05T19:34:55.000Z","updated_at":"2026-05-05T21:15:45.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/VedantJadhav701/Developer-Code-Intelligence-Agent","commit_stats":null,"previous_names":["vedantjadhav701/developer-code-intelligence-agent"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/VedantJadhav701/Developer-Code-Intelligence-Agent","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/VedantJadhav701%2FDeveloper-Code-Intelligence-Agent","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/VedantJadhav701%2FDeveloper-Code-Intelligence-Agent/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/VedantJadhav701%2FDeveloper-Code-Intelligence-Agent/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/VedantJadhav701%2FDeveloper-Code-Intelligence-Agent/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/VedantJadhav701","download_url":"https://codeload.github.com/VedantJadhav701/Developer-Code-Intelligence-Agent/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/VedantJadhav701%2FDeveloper-Code-Intelligence-Agent/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32713820,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-06T19:35:05.142Z","status":"ssl_error","status_checked_at":"2026-05-06T19:35:03.996Z","response_time":117,"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":["ai-agent","autonomous-agent","code-intelligence","coding-agent","developer-tools","local-llm","offline-ai","ollama","python","react-loop","self-review"],"created_at":"2026-05-06T22:01:55.147Z","updated_at":"2026-05-06T22:02:01.485Z","avatar_url":"https://github.com/VedantJadhav701.png","language":"Python","funding_links":["https://github.com/sponsors/VedantJadhav701"],"categories":[],"sub_categories":[],"readme":"# 🧠 DevAgent\n### Execution-Grounded Orchestration for Autonomous Local Coding.\n\n![DevAgent Showcase](readme-image.png)\n\n[**🌐 Live Site**](https://devagent-cli.vercel.app/) • [**📦 PyPI**](https://pypi.org/project/devagent-cli/) • [**📜 Docs**](https://devagent-cli.vercel.app/docs)\n\n[![PyPI version](https://badge.fury.io/py/devagent-cli.svg)](https://badge.fury.io/py/devagent-cli)\n[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)\n[![Ollama](https://img.shields.io/badge/Ollama-Local%20LLM-black.svg?logo=ollama)](https://ollama.ai)\n[![Safety: Sandboxed](https://img.shields.io/badge/Safety-Sandboxed-success.svg)](https://devagent-cli.vercel.app/benchmarks)\n\n**DevAgent** is a research-grade, local-first coding agent runtime designed to bridge the gap between LLM-generated logic and production-ready execution integrity.\n\n[Quick Start](#-quick-start) • [Architecture](#-high-integrity-architecture) • [Safety \u0026 Containment](#-safety-first-containment) • [Benchmarks](#-empirical-validation) • [Troubleshooting](docs/troubleshooting.md)\n\n\u003c/div\u003e\n\n---\n\n## 🛡️ The Problem: The \"LLM-Execution Gap\"\nMost autonomous coding agents fail because they operate in a vacuum. They generate code that looks correct but fails at runtime due to **environment drift**, **dependency conflicts**, or **invalid execution assumptions**.\n\nDevAgent is **Execution-Grounded Orchestration**. It doesn't just guess code; it manages the entire lifecycle of a fix:\n1.  **Discovery**: Scans and maps the environment.\n2.  **Isolation**: Provisions a clean, sandboxed virtual environment.\n3.  **Repair**: Autonomously resolves missing dependencies.\n4.  **Validation**: Verifies patches against your actual test suite.\n\n---\n\n## 🏗️ High-Integrity Architecture\nDevAgent v3.4.1 implements a multi-layer orchestration stack designed for reliability over hype.\n\n```mermaid\ngraph TD\n    subgraph \"Orchestration Layer\"\n    CLI[DevAgent CLI] --\u003e Planner[Task Planner]\n    Planner --\u003e Retrieval[Hierarchical Retrieval: FAISS + Ripgrep]\n    Retrieval --\u003e Agent[ReAct Agent: Thought/Action/Observation]\n    end\n\n    subgraph \"Execution Layer\"\n    Agent --\u003e Patch[Surgical Patch Engine: Line-Level Diffs]\n    Patch --\u003e Reviewer[Self-Review Loop: APPROVED/REVISE]\n    Reviewer --\u003e Runtime[Environment Runtime]\n    end\n\n    subgraph \"The Maturity Layer\"\n    Runtime --\u003e Discovery[Auto-Dependency Discovery]\n    Discovery --\u003e Isolation[Venv Isolation: .tmp_envs]\n    Isolation --\u003e Repair[Autonomous Repair Loop: repair_environment]\n    Repair --\u003e Validation[Validation Engine: pytest]\n    end\n\n    Validation -- Pass --\u003e Apply[Apply to Root Repo]\n    Validation -- Fail --\u003e Agent\n```\n\n---\n\n## 🚀 Quick Start\n\n### 1. Installation\nInstall the CLI via PyPI. Ensure [Ollama](https://ollama.ai) is running locally.\n\n```bash\npip install devagent-cli\n```\n\n### 2. Verify Infrastructure\nCheck your local environment, connectivity, and dependency health.\n```bash\ndevagent doctor\n```\n\n### 3. Run Your First Task\nExecute an autonomous fix on any repository.\n```bash\ndevagent run --task \"Implement input validation for the user login\" --root ./my-project\n```\n\n---\n\n## ✨ Advanced Features\n\n### 🔍 Hierarchical Retrieval\nInstead of dumping your entire codebase into a context window, DevAgent uses a multi-tier search:\n- **Global Map**: Scans the file structure to identify relevant modules.\n- **Semantic Tier**: FAISS-powered vector search for conceptual matching.\n- **Precision Tier**: Ripgrep for exact symbol/error discovery.\n\n### 🏖️ Environment Isolation \u0026 Repair\nDevAgent is the first local agent to treat the environment as a first-class citizen. It detects `requirements.txt` or `pyproject.toml`, creates an isolated `.tmp_envs/` runtime, and **autonomously installs missing packages** if it encounters a `ModuleNotFoundError`.\n\n### 🩹 Surgical Patch Engine\nMost agents ruin git history by rewriting entire files. DevAgent generates **line-level unified diffs**, applying only the necessary changes while preserving your code style, comments, and structure.\n\n---\n\n## 🔐 Safety-First Containment\nWe built DevAgent for engineers who care about their host systems.\n- **Dry-Run Mode**: Visualize every change before it happens.\n- **Atomic Snapshots**: A safety restore point is created before every execution.\n- **Instant Rollback**: Revert any agent intervention with `devagent rollback`.\n- **Sandbox Isolation**: Every run is contained in a separate workspace until validation passes 100%.\n\n---\n\n## 📊 Empirical Validation\nWe don't fake our success rates. DevAgent is evaluated against a public, messy benchmark suite.\n\n| Metric | Result | Infrastructure Status |\n| :--- | :--- | :--- |\n| **Dependency Repair** | 95% | ✅ Production Ready |\n| **Unit Bugfixes** | 80% | ✅ Highly Reliable |\n| **Refactoring** | 20% | 📈 Improving (Model Bounded) |\n| **Isolation Safety** | 100% | ✅ Absolute Containment |\n\n\u003e **Full Report**: [v3.4.1 Benchmark Analysis](docs/benchmarks.md)\n\n---\n\n## 🤝 Contributing\nBuilt with a focus on **Systems Thinking**. PRs that improve orchestration reliability, environment detection, or patch precision are highly encouraged.\n\n```bash\n# Clone and Install in Editable Mode\ngit clone https://github.com/VedantJadhav701/Developer-Code-Intelligence-Agent.git\npip install -e .\n```\n\n---\n\n\u003cdiv align=\"center\"\u003e\n**DevAgent — Local-First. Execution-Grounded. Infrastructure-Grade.**\n\u003c/div\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvedantjadhav701%2Fdeveloper-code-intelligence-agent","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fvedantjadhav701%2Fdeveloper-code-intelligence-agent","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvedantjadhav701%2Fdeveloper-code-intelligence-agent/lists"}