[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-80713":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":8,"htmlUrl":8,"language":9,"languages":8,"totalLinesOfCode":8,"stars":10,"forks":11,"watchers":12,"openIssues":13,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":14,"stars7d":15,"stars30d":16,"stars90d":14,"forks30d":14,"starsTrendScore":17,"compositeScore":18,"rankGlobal":8,"rankLanguage":8,"license":8,"archived":19,"fork":19,"defaultBranch":20,"hasWiki":21,"hasPages":19,"topics":22,"createdAt":8,"pushedAt":8,"updatedAt":23,"readmeContent":24,"aiSummary":25,"trendingCount":14,"starSnapshotCount":14,"syncStatus":26,"lastSyncTime":27,"discoverSource":28},80713,"local-ai-work-agent-template","clean-build-studio\u002Flocal-ai-work-agent-template","clean-build-studio",null,"Python",54,21,7,1,0,9,10,3,4.03,false,"main",true,[],"2026-06-12 02:04:05","# Setup\n\n## 1. Install Ollama\n\nThis project uses Ollama to run a local AI model on your computer.\n\nInstall Ollama from:\n\n```text\nhttps:\u002F\u002Follama.com\n```\n\nOn Linux, you can install Ollama with:\n\n```bash\ncurl -fsSL https:\u002F\u002Follama.com\u002Finstall.sh | sh\n```\n\nCheck that Ollama installed:\n\n```bash\nollama --version\n```\n\nIf Ollama is not running, start it with:\n\n```bash\nsudo systemctl start ollama\n```\n\nOptional status check:\n\n```bash\nsudo systemctl status ollama\n```\n\n---\n\n## 2. Pull the local model\n\nThis template uses Qwen3 4B by default:\n\n```bash\nollama pull qwen3:4b\n```\n\nYou can test the model directly with:\n\n```bash\nollama run qwen3:4b\n```\n\nThen try a simple prompt:\n\n```text\nSay hello in one sentence and confirm the local model is working.\n```\n\nTo exit the Ollama chat, type:\n\n```text\n\u002Fbye\n```\n\n---\n\n## 3. Clone this repo\n\nClone the project from GitHub:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002FYOUR-USERNAME\u002Flocal-ai-work-agent-template.git\n```\n\nMove into the project folder:\n\n```bash\ncd local-ai-work-agent-template\n```\n\nReplace `YOUR-USERNAME` with your actual GitHub username or organization name.\n\n---\n\n## 4. Create a Python virtual environment\n\nCreate the virtual environment:\n\n```bash\npython3 -m venv .venv\n```\n\nActivate it on Linux or macOS:\n\n```bash\nsource .venv\u002Fbin\u002Factivate\n```\n\nActivate it on Windows PowerShell:\n\n```powershell\n.venv\\Scripts\\Activate.ps1\n```\n\nWhen the virtual environment is active, you should see something like this at the beginning of your terminal line:\n\n```text\n(.venv)\n```\n\n---\n\n## 5. Install Python packages\n\nInstall the required packages:\n\n```bash\npip install -r requirements.txt\n```\n\nFor this first version, the project uses:\n\n```text\nrequests\npython-dotenv\n```\n\n`requests` lets Python send a request to Ollama.\n\n`python-dotenv` lets Python read local settings from the `.env` file.\n\n---\n\n## 6. Create your local `.env` file\n\nCopy the example environment file:\n\n```bash\ncp .env.example .env\n```\n\nYour `.env` file should include:\n\n```bash\nOLLAMA_MODEL=qwen3:4b\nOLLAMA_BASE_URL=http:\u002F\u002Flocalhost:11434\nAGENT_NAME=Local Work Agent\n\nENABLE_THINKING=false\nENABLE_FILE_ACCESS=true\nENABLE_CALENDAR_ACCESS=false\nENABLE_WEATHER_ACCESS=false\n\nDATA_INBOX_PATH=data\u002Finbox\nDATA_PROCESSED_PATH=data\u002Fprocessed\nDATA_OUTPUTS_PATH=data\u002Foutputs\n\nALLOW_FULL_COMPUTER_ACCESS=false\nSAVE_OUTPUTS=true\n```\n\nThe `.env.example` file is safe to commit to GitHub.\n\nThe real `.env` file should stay on your computer and should not be pushed to GitHub.\n\n---\n\n## 7. Run the first agent test\n\nFrom the project root, with your virtual environment activated, run:\n\n```bash\npython3 -m app.main\n```\n\nExpected output:\n\n```text\nLocal Work Agent is starting...\nUsing local model: qwen3:4b\nThinking enabled: False\n\nAgent response:\nLocal model connected.\n```\n\nThis proves the basic local path is working:\n\n```text\nPython project\n→ Ollama running locally\n→ Qwen model\n→ response back to Python\n```\n\n---\n\n# Troubleshooting\n\n## `python` command not found\n\nOn some Linux systems, use `python3` instead of `python`.\n\n```bash\npython3 -m venv .venv\n```\n\n---\n\n## Missing `venv`\n\nIf you see an error about `venv` not being available, install it:\n\n```bash\nsudo apt-get update\nsudo apt-get install python3-venv\n```\n\nThen try again:\n\n```bash\npython3 -m venv .venv\n```\n\n---\n\n## Missing `requests`\n\nIf you see:\n\n```text\nModuleNotFoundError: No module named 'requests'\n```\n\nMake sure your virtual environment is activated:\n\n```bash\nsource .venv\u002Fbin\u002Factivate\n```\n\nThen install dependencies:\n\n```bash\npip install -r requirements.txt\n```\n\n---\n\n## Ollama install error: `zstd`\n\nIf you see:\n\n```text\nERROR: This version requires zstd for extraction.\n```\n\nInstall `zstd` first:\n\n```bash\nsudo apt-get update\nsudo apt-get install zstd\n```\n\nThen rerun the Ollama install command:\n\n```bash\ncurl -fsSL https:\u002F\u002Follama.com\u002Finstall.sh | sh\n```\n\n---\n\n## Qwen overthinking a simple prompt\n\nQwen3 can sometimes generate reasoning text even for simple prompts.\n\nFor this first connection test, the template uses several guardrails:\n\n- `ENABLE_THINKING=false`\n- `\u002Fno_think`\n- a strict system prompt\n- low randomness settings\n- a response length limit\n- cleanup logic for leaked thinking text\n\nThe goal of the first test is not deep reasoning.\n\nThe goal is to confirm that Python can talk to Ollama and return a clean local response.\n\n---\n\n# Required files\n\nMake sure your `requirements.txt` has:\n\n```txt\nrequests\npython-dotenv\n```\n\nMake sure your `.gitignore` includes:\n\n```gitignore\n.env\n.venv\u002F\n\ndata\u002Finbox\u002F*\ndata\u002Fprocessed\u002F*\ndata\u002Foutputs\u002F*\n\n!data\u002Finbox\u002F.gitkeep\n!data\u002Fprocessed\u002F.gitkeep\n!data\u002Foutputs\u002F.gitkeep\n```","这个项目是一个用于在本地计算机上运行AI模型的工作代理模板。它基于Python语言，通过Ollama工具来管理并运行指定的AI模型，如默认配置中的Qwen3 4B版本。用户可以通过创建虚拟环境、安装必要的Python库（如requests和python-dotenv）以及设置环境变量文件来配置自己的工作代理。此外，该模板还支持对文件访问等特定功能进行开关控制。适合于需要在本地环境中利用AI模型处理任务但又希望保持数据隐私性和减少对外部服务依赖性的开发者或团队使用。",2,"2026-06-11 04:01:43","CREATED_QUERY"]