{"id":28958846,"url":"https://github.com/ibz-04/devseeker","last_synced_at":"2025-06-23T23:32:52.249Z","repository":{"id":294090727,"uuid":"979326675","full_name":"iBz-04/Devseeker","owner":"iBz-04","description":"Lightweight cli coding agent","archived":false,"fork":false,"pushed_at":"2025-05-18T21:30:39.000Z","size":16063,"stargazers_count":53,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-06-20T14:58:06.660Z","etag":null,"topics":["agentic-ai","ai-agent","aider","autonomous-agents","claude-code","cli-agent","coding-agents","coding-assistant","copilot","cursor-ai","langchain","llm-agents","python"],"latest_commit_sha":null,"homepage":"https://pypi.org/project/devseeker","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/iBz-04.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":"citation.cff","codeowners":".github/CODEOWNERS","security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-05-07T10:45:46.000Z","updated_at":"2025-06-16T22:41:30.000Z","dependencies_parsed_at":"2025-05-18T22:49:17.260Z","dependency_job_id":null,"html_url":"https://github.com/iBz-04/Devseeker","commit_stats":null,"previous_names":["ibz-04/devseeker"],"tags_count":6,"template":false,"template_full_name":null,"purl":"pkg:github/iBz-04/Devseeker","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iBz-04%2FDevseeker","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iBz-04%2FDevseeker/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iBz-04%2FDevseeker/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iBz-04%2FDevseeker/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/iBz-04","download_url":"https://codeload.github.com/iBz-04/Devseeker/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/iBz-04%2FDevseeker/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":261575698,"owners_count":23179556,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","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":["agentic-ai","ai-agent","aider","autonomous-agents","claude-code","cli-agent","coding-agents","coding-assistant","copilot","cursor-ai","langchain","llm-agents","python"],"created_at":"2025-06-23T23:31:41.186Z","updated_at":"2025-06-23T23:32:52.234Z","avatar_url":"https://github.com/iBz-04.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n   \u003ch2\u003eDevseeker\u003c/h2\u003e\n\u003c/div\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"https://res.cloudinary.com/diekemzs9/image/upload/v1746009112/extension_icon_1024px_jedbgf.png\" alt=\"Seeker-o1 Logo\" width=\"170\"/\u003e\n\u003c/p\u003e\n\n\u003cdiv align=\"center\"\u003e\n   \u003cp\u003e\u003cb\u003eDevSeeker is a coding agent inspired by claude code and codex\u003c/b\u003e\u003c/p\u003e\n\u003c/div\u003e\n\n\n\n\n [![PyPI version](https://img.shields.io/pypi/v/devseeker)](https://pypi.org/project/devseeker)\n\n\n\n\n**devseeker can :**\n- manage code files and folders\n- Create and edit code on command \n- Store code in short-term memory\n- review code changes\n- run code files \n- calculate token usage\n- offer multiple coding modes\n- support wide range of ai models\n\n\n## a little demo\n\nhttps://github.com/user-attachments/assets/9b095f14-f376-429f-85e3-4fe42522cec4\n\n**rare video of me losing a little game devseeker made :(**\n\n## Table of Contents\n\n- [Installation](#installation)\n- [Configuration](#configuration)\n- [Basic Usage](#basic-usage)\n- [Operational Modes](#operational-modes)\n- [CLI Options Reference](#cli-options-reference)\n- [Environment Variables](#environment-variables)\n- [Using Alternative Models](#using-alternative-models)\n- [Project Configuration](#project-configuration)\n- [File Selection](#file-selection)\n- [Troubleshooting](#troubleshooting)\n- [Windows-Specific Instructions](#windows-specific-instructions)\n- [Contributing](#contributing)\n- [License](#license)\n\n## Installation\n\n### Via pip (Recommended)\n\n```sh\npip install devseeker\n```\n\n### Development Installation\n\n```sh\ngit clone https://github.com/iBz-04/devseeker.git\ncd devseeker\npoetry install\npoetry env activate\n```\n\n## Configuration\n\nDevSeeker requires an OpenAI API key. You can set it in three ways:\n\n### 1. Environment variable\n\n```sh\nexport OPENAI_API_KEY=your_api_key\n```\n\n### 2. .env file\n\nCreate a `.env` file in your project directory:\n\n```\nOPENAI_API_KEY=your_api_key\n```\n\n### 3. Custom configuration\n\nFor advanced configuration, create a `devseeker.toml` file in your project:\n\n```toml\n[run]\nbuild = \"npm run build\"\ntest = \"npm run test\"\nlint = \"quick-lint-js\"\n\n[paths]\nbase = \"./src\"  # base directory to operate in\n```\n\n## Basic Usage\n\n### Creating a New Project\n\n1. Create an empty directory for your project\n2. Inside the directory, create a file named `prompt` containing your instructions\n3. Run DevSeeker:\n\n```sh\ndevseeker projects/my-new-project\n```\n\nWhen you run this command, DevSeeker will:\n- Present a welcome interface\n- Read your prompt (or ask for one if not found)\n- Generate code files based on your description\n- Create an entrypoint file for running the project\n- Ask if you want to execute the generated code\n\n### Improving Existing Code\n\n```sh\ndevseeker projects/my-existing-project -i\n```\n\nWhen you run DevSeeker in improve mode with the `-i` flag, it provides an interactive terminal UI that allows you to:\n\n1. Describe how you want to improve your application through natural language prompts\n2. Select which files should be modified (through an interactive file selection interface)\n3. Review proposed changes in a diff view (showing what will be added/removed)\n4. Accept or reject the changes before they're applied to your codebase\n\nYou can also use the `--skip-file-selection` or `-s` flag to bypass the interactive file selection:\n\n```sh\ndevseeker projects/my-existing-project -i -s\n```\n\n## Operational Modes\n\nDevSeeker supports several operational modes that change how it processes your prompts and generates code:\n\n### Standard Mode (Default)\n\nGenerates complete projects following your prompt.\n\n```sh\ndevseeker projects/my-project\n```\n\n### Improve Mode\n\nModifies existing code according to your instructions.\n\n```sh\ndevseeker projects/my-project -i\n```\n\n### Clarify Mode\n\nDiscusses specifications with you before implementing them.\n\n```sh\ndevseeker projects/my-project -c\n```\n\n### Lite Mode\n\nGenerates code using only your main prompt, without additional steps.\n\n```sh\ndevseeker projects/my-project -l\n```\n\n### Self-Heal Mode\n\nAutomatically fixes code when it fails during execution.\n\n```sh\ndevseeker projects/my-project -sh\n```\n\n## CLI Options Reference\n\nDevSeeker offers numerous command-line options to customize its behavior:\n\n| Option | Short | Default | Description |\n|--------|-------|---------|-------------|\n| `--model`, | `-m` | gpt-4o | The AI model to use |\n| `--temperature` | `-t` | 0.1 | Controls randomness in outputs (0.0-1.0) |\n| `--improve` | `-i` | False | Improves existing project |\n| `--lite` | `-l` | False | Runs using only the main prompt |\n| `--clarify` | `-c` | False | Discusses specifications before implementation |\n| `--self-heal` | `-sh` | False | Auto-fixes failing code |\n| `--azure` | `-a` | \"\" | Azure OpenAI endpoint URL |\n| `--use-custom-preprompts` | | False | Uses custom prompts from project workspace |\n| `--llm-via-clipboard` | | False | Uses clipboard for AI communication |\n| `--verbose` | `-v` | False | Enables verbose logging |\n| `--debug` | `-d` | False | Enables debug mode |\n| `--prompt_file` | | \"prompt\" | Path to text file with prompt |\n| `--entrypoint_prompt` | | \"\" | Path to file with entrypoint requirements |\n| `--image_directory` | | \"\" | Path to folder with images |\n| `--use_cache` | | False | Caches LLM responses to save tokens |\n| `--skip-file-selection` | `-s` | False | Skips interactive file selection in improve mode |\n| `--no_execution` | | False | Runs setup without calling LLM or writing code |\n| `--sysinfo` | | False | Outputs system information for debugging |\n| `--diff_timeout` | | 3 | Timeout for diff regexp searches |\n| `--help` | `-h` | | Shows help information |\n\n### Common Command Examples\n\n```sh\n# Basic usage - create a project from prompt\ndevseeker projects/my-project\n\n# Create a project with a specific model\ndevseeker projects/my-project -m gpt-4-turbo\n\n# Improve an existing project\ndevseeker projects/my-existing-project -i\n\n# Improve a project with higher temperature for more creative outputs\ndevseeker projects/my-project -i -t 0.5\n\n# Clarify requirements before implementation\ndevseeker projects/my-project -c\n\n# Use lite mode for faster generation\ndevseeker projects/my-project -l\n\n# Enable self-healing for auto-fixing code\ndevseeker projects/my-project -sh\n\n# Use Azure OpenAI service\ndevseeker projects/my-project --azure https://\u003cyour-resource-name\u003e.openai.azure.com\n\n# Display help information\ndevseeker --help\n\n# Display system information for troubleshooting\ndevseeker --sysinfo\n\n# Skip file selection in improve mode\ndevseeker projects/my-project -i -s\n\n# Use a specific prompt file\ndevseeker projects/my-project --prompt_file custom_prompt.txt\n\n# Use images in your prompt\ndevseeker projects/my-project --image_directory images/\n\n# Use custom preprompts\ndevseeker projects/my-project --use-custom-preprompts\n\n# Enable verbose logging\ndevseeker projects/my-project -v\n```\n\n## Environment Variables\n\nDevSeeker recognizes these environment variables:\n\n| Variable | Description |\n|----------|-------------|\n| `OPENAI_API_KEY` | Your OpenAI API key |\n| `MODEL_NAME` | Default model to use (e.g., \"gpt-4o\") |\n| `OPENAI_API_BASE` | Alternative API endpoint |\n| `LOCAL_MODEL` | Set to \"true\" when using local models |\n| `EDITOR` | Your preferred text editor |\n| `LANGCHAIN_WANDB_TRACING` | Enable W\u0026B tracing (set to \"true\") |\n| `WANDB_API_KEY` | Weights \u0026 Biases API key |\n\n## Using Alternative Models\n\n### Local Models with llama.cpp\n\n```bash\nexport OPENAI_API_BASE=\"http://localhost:8000/v1\"\nexport OPENAI_API_KEY=\"sk-your_local_key\"\nexport MODEL_NAME=\"CodeLlama\"\nexport LOCAL_MODEL=true\n```\n\n### Azure OpenAI\n\n```bash\ndevseeker --azure https://\u003cyour-resource-name\u003e.openai.azure.com my-project\n```\n\nSee [docs/open_models.md](docs/open_models.md) for detailed instructions.\n\n## Project Configuration\n\nDevSeeker can be configured with a `devseeker.toml` file in your project root:\n\n```toml\n[run]\nbuild = \"npm run build\"\ntest = \"npm run test\"\nlint = \"quick-lint-js\"\n\n[paths]\nbase = \"./frontend\"  # base directory for monorepos\nsrc = \"./src\"        # source directory for context\n\n[devseeker-app]      # used for devseeker.app integration\nproject_id = \"...\"\n```\n\n## File Selection\n\nWhen improving code, DevSeeker needs to know which files to include in its context. The file selection process:\n\n1. DevSeeker scans your project directory\n2. Creates a TOML file with file paths\n3. Opens this file in your text editor\n4. You uncomment lines for files you want to include\n5. Save and close the file to continue\n\nThe selection interface supports:\n- Color-coded file types\n- Intelligent defaults based on language\n- Filtering of common directories like `node_modules`\n\n## Troubleshooting\n\n### Common Issues\n\n#### API Key Not Found\n```\nError: OpenAI API key not found\n```\nSolution: Set your `OPENAI_API_KEY` as described in Configuration.\n\n#### Token Limit Exceeded\n```\nError: This model's maximum context length is exceeded\n```\nSolution: Select fewer files in improve mode or use a model with higher token limits.\n\n#### Execution Errors\nIf generated code fails to run, try:\n- Using self-heal mode: `devseeker path/to/project -sh`\n- Checking dependency installation\n- Inspecting generated logs in the project's `.devseeker/logs` directory\n\n## Windows-Specific Instructions\n\nWindows users should consult [WINDOWS_README.md](WINDOWS_README.md) for platform-specific details.\n\nKey differences:\n- Use `set` instead of `export` for environment variables\n- Path separators use backslashes\n- Some commands may require PowerShell\n\n## Contributing\n\nContributions are welcome! Please feel free to submit a Pull Request.\n\n## License\n\nThis project is licensed under the MIT License. See [LICENSE](LICENSE) for details.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fibz-04%2Fdevseeker","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fibz-04%2Fdevseeker","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fibz-04%2Fdevseeker/lists"}