[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-76092":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":16,"stars30d":17,"stars90d":16,"forks30d":16,"starsTrendScore":16,"compositeScore":18,"rankGlobal":10,"rankLanguage":10,"license":19,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":22,"hasPages":20,"topics":23,"createdAt":10,"pushedAt":10,"updatedAt":32,"readmeContent":33,"aiSummary":34,"trendingCount":16,"starSnapshotCount":16,"syncStatus":35,"lastSyncTime":36,"discoverSource":37},76092,"litellm-agent-platform","BerriAI\u002Flitellm-agent-platform","BerriAI","Self-hosted platform for running coding agents (Claude Code, Codex, Hermes) in isolated sandboxes with vault proxy. ","https:\u002F\u002Fdocs.litellm-agent-platform.ai",null,"TypeScript",542,59,6,12,0,174,59.33,"MIT License",false,"main",true,[24,25,26,27,28,29,30,31],"agent-platform","agent-vault","claude-code","codex","codex-cli","litellm","litellm-ai-gateway","sandbox","2026-06-12 04:01:20","# LiteLLM Agent Platform\n\n[![Discord](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-Chat-5865F2?logo=discord&logoColor=white)](https:\u002F\u002Fdiscord.gg\u002FNkxw3rm3EE)\n\nLiteLLM Agent Platform is self-hosted infrastructure for running coding agents — Claude Code, Codex, Hermes anything — inside isolated sandboxes with a credential vault, so agents can run with bypass-permissions on without ever seeing your real keys. Use it from the `lap` CLI in your terminal, the web UI, or call the API directly.\n\n**Learn more in the [docs](https:\u002F\u002Fdocs.litellm-agent-platform.ai\u002F)**.\n\n\u003Cimg width=\"964\" height=\"720\" alt=\"lap_2\" src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fa68b21fa-20b6-4ddc-a011-1bc9e6a3b296\" \u002F>\n\n## Get started\n\n> [!NOTE]\n> The `lap` CLI talks to a running instance of LiteLLM Agent Platform. To self-host the platform itself, jump to [Self-hosting](#self-hosting).\n\n1. Install the `lap` CLI:\n\n    ```bash\n    git clone https:\u002F\u002Fgithub.com\u002FBerriAI\u002Flitellm-agent-platform.git\n    cd litellm-agent-platform\u002Fcli && npm install\n    ln -sf \"$PWD\u002Fbin\u002Flap.mjs\" ~\u002F.local\u002Fbin\u002Flap\n    ```\n\n2. Point it at your platform:\n\n    ```bash\n    lap login\n    ```\n\n3. Open a sandbox:\n\n    ```bash\n    lap claude-code-cli1\n    ```\n\nThat spins up a fresh Kubernetes pod running Claude Code, attaches your local terminal to its TTY over a WebSocket, and drops you straight into the agent. The pod's env contains only stub credentials (e.g. `GITHUB_TOKEN=stub_github_a8f1`); the vault swaps them for real keys on every outbound TLS connection. Press **Ctrl-D** to detach; the session stays alive for 24h. See [docs\u002Flap-cli.md](docs\u002Flap-cli.md) for the full CLI.\n\n## Demo Video\n\n**▶ [Demo: setting up codex and claude-code sandboxes](https:\u002F\u002Fwww.loom.com\u002Fshare\u002Fa88d525111b1445cb90db342ade09ebc)**  ·  ~5 min\n\nEnd-to-end walkthrough: create an agent, open a sandbox from the `lap` CLI, attach a local terminal, run codex \u002F claude-code inside.\n\n\n## Supported harnesses\n\n| Harness | Quickstart |\n|---|---|\n| Claude Code | [docs.litellm-agent-platform.ai\u002Fquickstart\u002Fclaude-code](https:\u002F\u002Fdocs.litellm-agent-platform.ai\u002Fquickstart\u002Fclaude-code) |\n| Codex | [docs.litellm-agent-platform.ai\u002Fquickstart\u002Fcodex](https:\u002F\u002Fdocs.litellm-agent-platform.ai\u002Fquickstart\u002Fcodex) |\n| Hermes | [docs.litellm-agent-platform.ai\u002Fquickstart\u002Fhermes](https:\u002F\u002Fdocs.litellm-agent-platform.ai\u002Fquickstart\u002Fhermes) |\n\n\n## Self-hosting\n\nSandboxes run on Kubernetes via the [kubernetes-sigs\u002Fagent-sandbox](https:\u002F\u002Fgithub.com\u002Fkubernetes-sigs\u002Fagent-sandbox) CRD. Local dev uses [kind](https:\u002F\u002Fkind.sigs.k8s.io\u002F).\n\nPrereqs: Docker Desktop, `kind`, `kubectl`, `helm`, a LiteLLM gateway URL.\n\n```bash\nbin\u002Fkind-up.sh\ndocker compose up\n```\n\n`bin\u002Fkind-up.sh` is idempotent — provisions a kind cluster `agent-sbx`, installs the agent-sandbox controller, and loads the harness image. `docker compose up` boots Postgres, runs the schema migration, and starts web (`:3000`) + worker.\n\nOpen [localhost:3000](http:\u002F\u002Flocalhost:3000) to create an agent. Then point `lap` at it and run through the steps above.\n\nArchitecture and tuning: [docs\u002Fk8s-backend.md](docs\u002Fk8s-backend.md).\n\n### Deploying to production\n\nRecommended path: AWS EKS for the sandbox cluster, Render for web + worker. See [`deploy\u002F`](deploy\u002F) — `bin\u002Feks-up.sh` provisions the cluster, the Render Blueprint at the top of [`deploy\u002Frender\u002FREADME.md`](deploy\u002Frender\u002FREADME.md) is one click.\n\n## Architecture\n\n\u003Cimg width=\"1997\" height=\"1219\" alt=\"Xnapper-2026-05-08-19 10 50\" src=\"https:\u002F\u002Fraw.githubusercontent.com\u002FBerriAI\u002Flitellm-docs\u002Fmain\u002Fstatic\u002Fimg\u002Flitellm_agent_platform_alpha.png\" \u002F>\n\n## Developer API\n\nCreate an agent, open a session, send a message, read the reply — directly with curl. See [`docs\u002Fspawn-task-agent.md`](docs\u002Fspawn-task-agent.md) and [`src\u002Fserver\u002FDEVELOPER.md`](src\u002Fserver\u002FDEVELOPER.md).\n\n## License\n\nMIT — see [LICENSE](LICENSE).\n# bump 1779140176\n","LiteLLM Agent Platform 是一个自托管平台，用于在隔离沙箱中运行编码代理（如Claude Code、Codex和Hermes），并通过凭证保险库确保安全。其核心功能包括使用Kubernetes管理沙箱环境，支持通过CLI、Web UI或直接调用API来操作，同时保证敏感信息不会泄露给代理。该平台采用TypeScript开发，具有良好的扩展性和安全性。适用于需要高度隔离和安全性的开发场景，例如涉及敏感数据处理的代码生成或自动化任务执行。",2,"2026-06-01 03:45:12","CREATED_QUERY"]