diff --git a/README.md b/README.md index 111eee9161..824bc5bddd 100644 --- a/README.md +++ b/README.md @@ -1,652 +1,175 @@ - - # codex-lb -**English** | [简体中文](./README.zh-CN.md) +[English](README.md) | [简体中文](README.zh-CN.md) + +A local proxy and load balancer for ChatGPT accounts. -Load balancer for ChatGPT accounts. Pool multiple accounts, track usage, manage API keys, view everything in a dashboard. +codex-lb pools accounts behind OpenAI-compatible endpoints, keeps a Codex-native endpoint for Codex clients, tracks usage, and provides a browser dashboard for account, key, and routing controls. -| ![dashboard](docs/screenshots/dashboard.jpg) | ![accounts](docs/screenshots/accounts.jpg) | -|:---:|:---:| +[![Fork activity](https://img.shields.io/github/last-commit/glopyglerky/codex-lb?label=fork%20updated)](https://github.com/glopyglerky/codex-lb/commits/main) +[![Upstream release](https://img.shields.io/github/v/release/Soju06/codex-lb?label=upstream%20release)](https://github.com/Soju06/codex-lb/releases/latest) +[![License](https://img.shields.io/github/license/Soju06/codex-lb)](LICENSE) -
-More screenshots +> This repository is a personal fork of [Soju06/codex-lb](https://github.com/Soju06/codex-lb). It does not publish its own release or container image. The quick start below runs the upstream image. Review this fork's changes before deploying its source. -| Settings | Login | -|:---:|:---:| -| ![settings](docs/screenshots/settings.jpg) | ![login](docs/screenshots/login.jpg) | +## What it does -| Dashboard (dark) | Accounts (dark) | Settings (dark) | -|:---:|:---:|:---:| -| ![dashboard-dark](docs/screenshots/dashboard-dark.jpg) | ![accounts-dark](docs/screenshots/accounts-dark.jpg) | ![settings-dark](docs/screenshots/settings-dark.jpg) | +- Balances requests across eligible ChatGPT accounts while preserving sticky conversation routing. +- Exposes `/v1` for OpenAI-compatible clients and `/backend-api/codex` for Codex-native traffic. +- Tracks account usage, request history, costs, quota resets, and routing decisions. +- Issues scoped API keys with model and rate limits. +- Supports SQLite by default and PostgreSQL for shared deployments. +- Provides dashboard password, TOTP, trusted-header, and explicitly disabled authentication modes. +- Runs locally, in Docker, or through the included Helm chart. -
+No routing strategy can guarantee account safety. Use normal request volumes, follow the applicable terms, and keep sticky routing enabled unless you have a specific reason to change it. -## Features +## How it works - - - - - - - - - - - -
Account Pooling
Load balance across multiple ChatGPT accounts		Usage Tracking
Per-account tokens, cost, 28-day trends		API Keys
Per-key rate limits by token, cost, window, model
Dashboard Auth
Password + optional TOTP		OpenAI-compatible
Codex CLI, OpenCode, any OpenAI client		Auto Model Sync
Available models fetched from upstream
+```text +client -> authentication -> model and account eligibility -> sticky routing + -> upstream Codex session -> streamed response -> usage settlement +``` -## Routing Strategy Guide +The routing contract, retry rules, and session ownership live in [OpenSpec](openspec/specs/account-routing/spec.md). The dashboard exposes the supported strategies. Capacity weighted and relative availability are the sensible defaults for most pools. -The dashboard setting **Routing strategy** controls how eligible accounts are selected for each request. No strategy can guarantee account-safety outcomes; conservative use still depends on staying within OpenAI terms, using normal request volumes, and avoiding traffic patterns that would be unusual for your accounts. +## Current status -For low-volume, policy-compliant personal use, start with **Capacity weighted** or **Relative availability** and keep sticky threads enabled. Those strategies preserve session locality while avoiding sudden all-traffic shifts to a single account. +The fork's default branch is active development code and currently identifies itself as an alpha package. It has no fork-specific release. Use the live badges above for fork activity and the latest upstream release, and use `GET /v1/models` for the current model catalog. Copied model lists age badly. -| Routing strategy | Behavior | Trade-offs and recommended use | -|---|---|---| -| Capacity weighted | Prefers accounts with more usable quota headroom. | Good default for mixed pools and normal compliant usage. | -| Relative availability | Draws from the strongest available accounts with configurable weighting. | Smooths distribution while still preferring healthier accounts. | -| Usage weighted | Reacts to observed recent usage. | Useful when usage history should influence selection, but less direct than capacity-based routing. | -| Round robin | Cycles evenly through eligible accounts. | Simple and predictable, but ignores quota shape and reset timing. | -| Fill first | Uses one account heavily before moving on. | Best for controlled drain tests; less conservative for everyday traffic. | -| Sequential drain | Drains accounts in a fixed order. | Useful for maintenance or explicit account rotation, not a normal safety-first default. | -| Reset drain | Prioritizes capacity near reset windows. | Helps consume expiring quota, but can create timing-shaped bursts. | -| Single account | Pins all traffic to one selected active account. | Useful for isolation and debugging; no load balancing. | +The normative behavior lives under [`openspec/specs/`](openspec/specs/). Configuration defaults come from [`.env.example`](.env.example). Those files are better authorities than a dated feature matrix in this README. -## Quick Start +## Quick start ```bash -# Docker (recommended) docker volume create codex-lb-data docker run -d --name codex-lb \ -p 2455:2455 -p 1455:1455 \ -v codex-lb-data:/var/lib/codex-lb \ ghcr.io/soju06/codex-lb:latest - -# or uvx -uvx codex-lb ``` -Open [localhost:2455](http://localhost:2455) → Add account → Done. - -## Remote Setup - -When accessing the dashboard remotely for the first time, a bootstrap token is required to set the initial password. - -**Auto-generated (default):** On first startup (no password configured), the server generates a one-time token and prints it to logs: +Open [localhost:2455](http://localhost:2455), add an account, then check the live model catalog: ```bash -docker logs codex-lb -# ============================================ -# Dashboard bootstrap token (first-run): -# -# ============================================ +curl http://127.0.0.1:2455/v1/models ``` -Open the dashboard → enter the token + new password → done. The token is shared across replicas and remains valid until a password is set. In multi-replica setups, replicas must share the same encryption key (the Helm chart default) for restart recovery to work. +You can also run the upstream Python package with `uvx codex-lb`. -**Manual token:** To use a fixed token instead, set the env var before starting: +### First remote login + +Local dashboard access bypasses first-run bootstrap. Remote access requires a one-time token when no password exists. The server prints the generated token at startup: ```bash -docker run -d --name codex-lb \ - -e CODEX_LB_DASHBOARD_BOOTSTRAP_TOKEN=your-secret-token \ - -p 2455:2455 -p 1455:1455 \ - -v codex-lb-data:/var/lib/codex-lb \ - ghcr.io/soju06/codex-lb:latest +docker logs codex-lb ``` -**Local access** (localhost) bypasses bootstrap entirely — no token needed. - -## Client Setup - -Point any OpenAI-compatible client at codex-lb. If [API key auth](#api-key-authentication) is enabled, pass a key from the dashboard as a Bearer token. - -Model availability is discovered from the upstream Codex model catalog and can vary by account plan, workspace, rollout, and upstream deprecation state. Prefer the live `GET /v1/models` or `GET /backend-api/codex/models` response over a copied static table when configuring clients or API-key model allowlists. +Set `CODEX_LB_DASHBOARD_BOOTSTRAP_TOKEN` before startup if you need a fixed token. Replicas must share the same encryption key so first-run and session state survive restarts. -| Logo | Client | Endpoint | Config | -|---|--------|----------|--------| -| OpenAI | **Codex CLI** | `http://127.0.0.1:2455/backend-api/codex` | `~/.codex/config.toml` | -| OpenCode | **OpenCode** | `http://127.0.0.1:2455/v1` | `~/.config/opencode/opencode.json` | -| OpenClaw | **OpenClaw** | `http://127.0.0.1:2455/v1` | `~/.openclaw/openclaw.json` | -| Python | **OpenAI Python SDK** | `http://127.0.0.1:2455/v1` | Code | +## Connect Codex -
-OpenAI Codex CLI / IDE Extension -
- -`~/.codex/config.toml`: +Pick a model from `/v1/models`, then add a provider to `~/.codex/config.toml`: ```toml -model = "gpt-5.3-codex" -model_reasoning_effort = "xhigh" +model = "MODEL_ID" model_provider = "codex-lb" -[model_providers.codex-lb] -name = "openai" # required — enables remote /responses/compact. Lowercase since Codex 2026-05-23; older "OpenAI" stops resolving gpt-5.5 -base_url = "http://127.0.0.1:2455/backend-api/codex" -wire_api = "responses" -supports_websockets = true -requires_openai_auth = true # required for codex app -``` - -This documented `requires_openai_auth = true` setup uses Codex-backed -authentication and does not need an `x-openai-actor-authorization` marker to be -eligible for Codex's built-in `$imagegen` tool. Provider configurations that -intentionally skip OpenAI login have a different eligibility path; see the -[Images compatibility context](openspec/specs/images-api-compat/context.md#codex-provider-eligibility). - -Optional: enable native upstream WebSockets for Codex streaming while keeping `codex-lb` pooling: - -```bash -export CODEX_LB_UPSTREAM_STREAM_TRANSPORT=websocket -``` - -`auto` is the default and uses native WebSockets for native Codex headers or models that prefer them. -You can also switch this in the dashboard under Settings -> Routing -> Upstream stream transport. - -Note: Codex itself does not currently expose a stable documented `wire_api = "websocket"` provider mode. -If you want to experiment on the Codex side, the current CLI exposes under-development feature flags: - -```toml -[features] -responses_websockets = true -# or -responses_websockets_v2 = true -``` - -These flags are experimental and do not replace `wire_api = "responses"`. - -Upstream websocket handshakes automatically honor standard proxy environment variables when they are -present. `wss://` handshakes check `wss_proxy`, `socks_proxy`, `https_proxy`, and `all_proxy`; -plain `ws://` handshakes also check `ws_proxy` and `http_proxy`. Set -`CODEX_LB_UPSTREAM_WEBSOCKET_TRUST_ENV=false` only when websocket handshakes must bypass those -environment proxies and connect directly. - -**With [API key auth](#api-key-authentication):** - -```toml [model_providers.codex-lb] name = "openai" base_url = "http://127.0.0.1:2455/backend-api/codex" wire_api = "responses" -env_key = "CODEX_LB_API_KEY" supports_websockets = true -requires_openai_auth = true # required for codex app -``` - -```bash -export CODEX_LB_API_KEY="sk-clb-..." # key from dashboard -codex +requires_openai_auth = true ``` -**Verify WebSocket transport** - -Use a one-off debug run: - -```bash -RUST_LOG=debug codex exec "Reply with OK only." -``` - -Healthy websocket signals: - -- CLI logs contain `connecting to websocket` and `successfully connected to websocket` -- `codex-lb` logs show `WebSocket /backend-api/codex/responses` -- `codex-lb` logs do **not** show fallback `POST /backend-api/codex/responses` for the same run - -If you run `codex-lb` behind a reverse proxy, make sure it forwards WebSocket upgrades. - -**Migrating from direct OpenAI** — `codex resume` filters by `model_provider`; -old sessions won't appear until you re-tag them. Use the built-in retag command -instead of editing Codex files by hand; see -[Codex session retagging](openspec/specs/runtime-portability/context.md#codex-session-retagging) for backups, Docker, WSL, -and rollback details. +If API key authentication is enabled, add `env_key = "CODEX_LB_API_KEY"` to the provider and export the key before starting Codex: ```bash -# Preview what will change first. -codex-lb codex-sessions retag --from openai --to codex-lb --dry-run - -# Then close Codex/Codex CLI and apply the retag. -codex-lb codex-sessions retag --from openai --to codex-lb --yes -``` - -
- -
-OpenCode OpenCode -
- -> **Important**: Use the built-in `openai` provider with `baseURL` override — not a custom provider with `@ai-sdk/openai-compatible`. Custom providers use the Chat Completions API which **drops reasoning/thinking content**. The built-in `openai` provider uses the Responses API, which properly preserves `encrypted_content` and multi-turn reasoning state. - -Before starting, please ensure that all existing OpenAI credentials is cleared in `~/.local/share/opencode/auth.json` -You can clean the config by using this one-liner -`jq 'del(.openai)' ~/.local/share/opencode/auth.json > auth.json.tmp && mv auth.json.tmp ~/.local/share/opencode/auth.json` - -`~/.config/opencode/opencode.json`: - -```jsonc -{ - "$schema": "https://opencode.ai/config.json", - "provider": { - "openai": { - "options": { - "baseURL": "http://127.0.0.1:2455/v1", - "apiKey": "{env:CODEX_LB_API_KEY}" - }, - "models": { - "gpt-5.4": { - "name": "GPT-5.4", - "reasoning": true, - "options": { "reasoningEffort": "high", "reasoningSummary": "detailed" }, - "limit": { "context": 1050000, "output": 128000 } - }, - "gpt-5.3-codex": { - "name": "GPT-5.3 Codex", - "reasoning": true, - "options": { "reasoningEffort": "high", "reasoningSummary": "detailed" }, - "limit": { "context": 272000, "output": 65536 } - }, - "gpt-5.1-codex-mini": { - "name": "GPT-5.1 Codex Mini", - "reasoning": true, - "options": { "reasoningEffort": "high", "reasoningSummary": "detailed" }, - "limit": { "context": 272000, "output": 65536 } - }, - "gpt-5.3-codex-spark": { - "name": "GPT-5.3 Codex Spark", - "reasoning": true, - "options": { "reasoningEffort": "xhigh", "reasoningSummary": "detailed" }, - "limit": { "context": 128000, "output": 65536 } - } - } - } - }, - "model": "openai/gpt-5.3-codex" -} -``` - -This overrides the built-in `openai` provider's endpoint to point at codex-lb while keeping the Responses API code path that handles reasoning properly. - -```bash -export CODEX_LB_API_KEY="sk-clb-..." # key from dashboard -opencode -``` - -
- -
-OpenClaw OpenClaw -
- -`~/.openclaw/openclaw.json`: - -```jsonc -{ - "agents": { - "defaults": { - "model": { "primary": "codex-lb/gpt-5.4" }, - "models": { - "codex-lb/gpt-5.4": { "params": { "cacheRetention": "short" } } - "codex-lb/gpt-5.4-mini": { "params": { "cacheRetention": "short" } } - "codex-lb/gpt-5.3-codex": { "params": { "cacheRetention": "short" } } - } - } - }, - "models": { - "mode": "merge", - "providers": { - "codex-lb": { - "baseUrl": "http://127.0.0.1:2455/v1", - "apiKey": "${CODEX_LB_API_KEY}", // or "dummy" if API key auth is disabled - "api": "openai-responses", - "models": [ - { - "id": "gpt-5.4", - "name": "gpt-5.4 (codex-lb)", - "contextWindow": 1050000, - "contextTokens": 272000, - "maxTokens": 4096, - "input": ["text"], - "reasoning": false - }, - { - "id": "gpt-5.4-mini", - "name": "gpt-5.4-mini (codex-lb)", - "contextWindow": 400000, - "contextTokens": 272000, - "maxTokens": 4096, - "input": ["text"], - "reasoning": false - }, - { - "id": "gpt-5.3-codex", - "name": "gpt-5.3-codex (codex-lb)", - "contextWindow": 400000, - "contextTokens": 272000, - "maxTokens": 4096, - "input": ["text"], - "reasoning": false - } - ] - } - } - } -} -``` - -Set the env var or replace `${CODEX_LB_API_KEY}` with a key from the dashboard. If API key auth is disabled, -local requests can omit the key, but non-local requests are still rejected until proxy authentication is configured. - -The `/v1` route is the simplest OpenAI-compatible setup. If your OpenClaw build uses a Codex-native provider path such as `openai-codex-responses` and needs Codex-style usage/accounting behavior, point that provider at `http://127.0.0.1:2455/backend-api/codex` instead. For third-party Codex-compatible backends, the client must allow opaque bearer-token passthrough and should only send `chatgpt-account-id` when it actually decoded one from an official ChatGPT/Codex token. - -
- -
-Python OpenAI Python SDK -
- -```python -from openai import OpenAI - -client = OpenAI( - base_url="http://127.0.0.1:2455/v1", - api_key="sk-clb-...", # from dashboard, or any non-empty string if auth is disabled -) - -response = client.chat.completions.create( - model="gpt-5.3-codex", - messages=[{"role": "user", "content": "Hello!"}], -) -print(response.choices[0].message.content) +export CODEX_LB_API_KEY="sk-clb-..." +codex ``` -
+The same proxy also supports OpenCode, OpenClaw, and the OpenAI SDK. Point Responses API clients at `/v1`. Client behavior and edge cases are documented in the [Responses API context](openspec/specs/responses-api-compat/context.md), [chat completions context](openspec/specs/chat-completions-compat/context.md), and [runtime portability context](openspec/specs/runtime-portability/context.md). -## API Key Authentication +## Authentication -API key auth is **disabled by default**. In that mode, only local requests to the protected proxy routes can -proceed without a key; non-local requests are rejected until proxy authentication is configured. Enable it in -**Settings → API Key Auth** on the dashboard when clients connect remotely or through Docker, VM, or container -networking that appears non-local to the service. +Proxy API key authentication is disabled by default. Without it, protected proxy routes accept only local requests. Enable API key authentication in the dashboard before allowing remote clients, then send: -When enabled, clients must pass a valid API key as a Bearer token: - -``` +```text Authorization: Bearer sk-clb-... ``` -The protected proxy routes covered by this setting are: - -- `/v1/*` (except `/v1/usage`, which always requires a valid key) -- `/backend-api/codex/*` -- `/backend-api/transcribe` +Dashboard authentication has three explicit modes: -**Creating keys**: Dashboard → API Keys → Create. The full key is shown **only once** at creation. Keys support optional expiration, model restrictions, and rate limits (tokens / cost per day / week / month). +- `standard` uses a password and optional TOTP. +- `trusted_header` accepts an identity header only from configured proxy CIDRs. +- `disabled` removes application-level dashboard authentication and should sit behind a separate access boundary. -## Configuration +See the [admin authentication context](openspec/specs/admin-auth/context.md) and [API key contract](openspec/specs/api-keys/spec.md) before exposing the service beyond localhost. -Environment variables with `CODEX_LB_` prefix or `.env.local`. See [`.env.example`](.env.example). -SQLite is the default database backend; PostgreSQL is optional via `CODEX_LB_DATABASE_URL` (for example `postgresql+asyncpg://...`). +## Configuration and data -The Docker Compose `postgres` profile uses the Postgres 18 image and mounts the named data volume at -`/var/lib/postgresql`, the parent of the image's versioned `PGDATA` directory. +Environment variables use the `CODEX_LB_` prefix. [`.env.example`](.env.example) is the complete configuration reference. -Existing Postgres 16 compose volumes must be upgraded before the Postgres 18 container starts: - -```bash -docker compose --profile postgres stop postgres -docker run --rm -v codex-lb-postgres-data:/var/lib/postgresql -v "$PWD:/backup" alpine \ - tar -C /var/lib/postgresql -czf /backup/codex-lb-postgres-data-before-pg18.tgz . -docker compose --profile postgres-upgrade run --rm postgres-upgrade -docker compose --profile postgres up -d postgres -``` - -The `postgres-upgrade` profile runs `pg_upgrade` in one-shot mode against the same named volume and exits after the -data directory has been upgraded to the Postgres 18 layout. Because that helper mounts and rewrites the operator's -database volume, Compose pins the helper image by digest; refresh and review the digest deliberately when changing the -helper image tag. Keep the backup until the application has started and `codex-lb-db check` succeeds against the -upgraded database. - -The normal `postgres` service refuses to start when it detects the old root-level `PG_VERSION` file from a pre-18 -Compose volume. If that guard fires, run the `postgres-upgrade` profile above before starting Postgres again. -It also refuses nested `/var/lib/postgresql/data` directories that still report a pre-18 major version, because those -layouts need an explicit pg_upgrade before the Postgres 18 container can safely open them. - -### Dashboard authentication modes - -`codex-lb` supports three dashboard auth modes via environment variables: +| Runtime | Data path | +| --- | --- | +| Local or `uvx` | `~/.codex-lb/` | +| Docker | `/var/lib/codex-lb/` | -- `CODEX_LB_DASHBOARD_AUTH_MODE=standard` — built-in dashboard password with optional TOTP from the Settings page. -- `CODEX_LB_DASHBOARD_AUTH_MODE=trusted_header` — trust a reverse-proxy auth header such as Authelia's `Remote-User`, but only from `CODEX_LB_FIREWALL_TRUSTED_PROXY_CIDRS`. Built-in password/TOTP remain available as an optional fallback, and password/TOTP management still requires a fallback password session. -- `CODEX_LB_DASHBOARD_AUTH_MODE=disabled` — fully bypass dashboard auth. Use only behind network restrictions or external auth. Built-in password/TOTP management is disabled in this mode. +Back up that directory before upgrades. SQLite is the default. PostgreSQL setup and migration rules are in the [database context](openspec/specs/database-backends/context.md). -`trusted_header` mode also requires: +## Kubernetes ```bash -CODEX_LB_FIREWALL_TRUST_PROXY_HEADERS=true -CODEX_LB_FIREWALL_TRUSTED_PROXY_CIDRS=172.18.0.0/16 -CODEX_LB_DASHBOARD_AUTH_PROXY_HEADER=Remote-User +helm install codex-lb oci://ghcr.io/soju06/charts/codex-lb \ + --set postgresql.auth.password=changeme \ + --set config.databaseMigrateOnStartup=true \ + --set migration.schemaGate.enabled=false +kubectl port-forward svc/codex-lb 2455:2455 ``` -If the trusted header is missing and no fallback password is configured, the dashboard fails closed and shows a reverse-proxy-required message instead of loading the UI. +The [Helm chart README](deploy/helm/codex-lb/README.md) covers external databases, ingress, authentication, observability, and multi-replica routing. -### Docker examples +## Development -**Authelia / trusted header** +This project requires Python 3.13 and uses `uv` for the locked environment. ```bash -docker run -d --name codex-lb \ - -p 2455:2455 -p 1455:1455 \ - -e CODEX_LB_DASHBOARD_AUTH_MODE=trusted_header \ - -e CODEX_LB_DASHBOARD_AUTH_PROXY_HEADER=Remote-User \ - -e CODEX_LB_FIREWALL_TRUST_PROXY_HEADERS=true \ - -e CODEX_LB_FIREWALL_TRUSTED_PROXY_CIDRS=172.18.0.0/16 \ - -v codex-lb-data:/var/lib/codex-lb \ - ghcr.io/soju06/codex-lb:latest -``` - -**Hard override / no app-level dashboard auth** +uv sync --frozen +cd frontend && bun install && cd .. -```bash -docker run -d --name codex-lb \ - -p 2455:2455 -p 1455:1455 \ - -e CODEX_LB_DASHBOARD_AUTH_MODE=disabled \ - -v codex-lb-data:/var/lib/codex-lb \ - ghcr.io/soju06/codex-lb:latest +uv run fastapi run app/main.py --reload +cd frontend && bun run dev ``` -For Helm, pass the same values through `extraEnv`. - -## Data - -| Environment | Path | -|-------------|------| -| Local / uvx | `~/.codex-lb/` | -| Docker | `/var/lib/codex-lb/` | - -Backup this directory to preserve your data. - -## Troubleshooting - -- [Usage and quota - why does codex-lb still say `rate_limited` when Codex Desktop says reset?](openspec/specs/usage-refresh-policy/context.md) - -## Kubernetes +Run the focused checks before opening a change: ```bash -helm install codex-lb oci://ghcr.io/soju06/charts/codex-lb \ - --set postgresql.auth.password=changeme \ - --set config.databaseMigrateOnStartup=true \ - --set migration.schemaGate.enabled=false -kubectl port-forward svc/codex-lb 2455:2455 +uv run --frozen ruff check app tests +uv run --frozen ruff format --check app tests +uv run --frozen ty check app +uv run --frozen pytest tests/unit -q ``` -Open [localhost:2455](http://localhost:2455) → Add account → Done. +Behavior changes start in [`openspec/`](openspec/). Read [the contribution rules](.github/CONTRIBUTING.md) for the repository's merge gates. -The Helm chart auto-configures HTTP `/responses` owner handoff for multi-replica installs using a headless-service DNS name per pod. The default cluster domain is `cluster.local`; set Helm `clusterDomain` if your cluster uses a different suffix. Override `config.sessionBridgeAdvertiseBaseUrl` only if pods must be reached through a different internal address. +## Repository map -For external database, production config, ingress, observability, and more see the [Helm chart README](deploy/helm/codex-lb/README.md). +| Path | Contents | +| --- | --- | +| [`app/`](app/) | FastAPI service, routing, persistence, authentication, and proxy runtime | +| [`frontend/`](frontend/) | Dashboard application | +| [`openspec/`](openspec/) | Normative behavior and operating context | +| [`deploy/helm/codex-lb/`](deploy/helm/codex-lb/) | Kubernetes chart and deployment guide | +| [`tests/`](tests/) | Unit, integration, compatibility, and release checks | -Fast Mode and service-tier behavior is documented in [Responses API compatibility context](openspec/specs/responses-api-compat/context.md#fast-mode-and-service-tiers). +## Contributing and security -## Development +This fork retains the upstream project's history and contributors. See the [contributor graph](https://github.com/Soju06/codex-lb/graphs/contributors) and [upstream contribution guide](https://github.com/Soju06/codex-lb/blob/main/.github/CONTRIBUTING.md). -```bash -# Docker -docker compose watch +Report vulnerabilities through the [security policy](.github/SECURITY.md). Do not post secrets, account exports, access tokens, or production logs in public issues. -# Local -uv sync && cd frontend && bun install && cd .. -uv run fastapi run app/main.py --reload # backend :2455 -cd frontend && bun run dev # frontend :5173 -``` +## License -## Contributors ✨ - -Thanks goes to these wonderful people ([emoji key](https://allcontributors.org/en/reference/emoji-key/)): - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Soju06
Soju06
💻 ⚠️ 🚧 🚇		Hwanmoo Yong
Hwanmoo Yong
💻 ⚠️		Jonas Kamsker
Jonas Kamsker
💻 🐛 🚧		Quack
Quack
💻 🐛 🚧 🎨		Jill Kok, San Mou
Jill Kok, San Mou
💻 ⚠️ 🚧 🐛		PARK CHANYOUNG
PARK CHANYOUNG
📖 💻 ⚠️		Choi138
Choi138
💻 🐛 ⚠️
LYA⚚CAP⚚OCEAN
LYA⚚CAP⚚OCEAN
💻 ⚠️		Diógenes Castro
Diógenes Castro
💻 ⚠️		Eugene Korekin
Eugene Korekin
💻 🐛 ⚠️		jordan
jordan
💻 🐛 ⚠️		DOCaCola
DOCaCola
🐛 ⚠️ 📖		JoeBlack2k
JoeBlack2k
💻 🐛 ⚠️		Peter A.
Peter A.
📖 💻 🐛
Hannah Markfort
Hannah Markfort
💻 ⚠️		mws-weekend-projects
mws-weekend-projects
💻 ⚠️		Quang Do
Quang Do
💻 ⚠️		Anand Aiyer
Anand Aiyer
🐛 💻 ⚠️		defin85
defin85
💻 🐛 ⚠️		Jacky Fong
Jacky Fong
💻 🐛 💬 🚧 ⚠️		flokosti96
flokosti96
💻 ⚠️
Woonggi Min
Woonggi Min
💻 ⚠️		Yigit Konur
Yigit Konur
🐛 💻		Ruben
Ruben
💻 ⚠️ 🐛		Steve Santacroce
Steve Santacroce
💻 ⚠️ 🐛		Hugh Do
Hugh Do
💻 ⚠️		Hubert Salwin
Hubert Salwin
💻 ⚠️		Teemu Koskinen
Teemu Koskinen
📖
Yu Peng Zheng
Yu Peng Zheng
📖 💻		embogomolov
embogomolov
💻 ⚠️		Renat Sharipov
Renat Sharipov
💻 ⚠️		Liu Rui
Liu Rui
📖 💻 ⚠️ 🐛		OverHash
OverHash
💻 ⚠️		Kazet
Kazet
💻 ⚠️		Bala Kumar
Bala Kumar
💻 ⚠️ 🤔
ihazgithub
ihazgithub
💻 ⚠️		Temirkhan
Temirkhan
💻 ⚠️ 📖 🐛		tobwen
tobwen
💻 ⚠️ 🐛		Rio
Rio
💻 🐛 ⚠️		Mika
Mika
💻 📖 ⚠️		Darafei Praliaskouski
Darafei Praliaskouski
💻 📖 ⚠️ 🐛		Maxim Feofilov
Maxim Feofilov
💻 ⚠️
JeffKandt
JeffKandt
⚠️ 👀		klaascommerce
klaascommerce
💻 ⚠️		ozpool
ozpool
🤔 📖 💻 ⚠️		Manu
Manu
⚠️ 👀		Wojtek Majewski
Wojtek Majewski
⚠️		Andrew Noble
Andrew Noble
💻 ⚠️		Josu Gorostegui
Josu Gorostegui
💻 ⚠️
Linus Mixson
Linus Mixson
💻 ⚠️		Lotfree
Lotfree
💻 ⚠️ 📖 🐛		timefox
timefox
💻 ⚠️		Nikhil
Nikhil
💻 ⚠️		Miha Orazem
Miha Orazem
💻 ⚠️		Steven (Minh) Dang
Steven (Minh) Dang
📖		onlysdesign-ui
onlysdesign-ui
💻 ⚠️
Mahir Taha Özdin
Mahir Taha Özdin
🤔 💻 ⚠️		hikki
hikki
💻 🎨		Nataprom
Nataprom
💻 ⚠️		Iweisc
Iweisc
💻 ⚠️		ram/haidar
ram/haidar
💻		Rudra Tiwari
Rudra Tiwari
💻		Wu Chao
Wu Chao
💻
zwd0313
zwd0313
💻		jhordanjw123
jhordanjw123
💻 ⚠️		mastertyko
mastertyko
💻 ⚠️		NeoClaw
NeoClaw
💻		abarsegov
abarsegov
💻		Akshay Kakatkar
Akshay Kakatkar
💻 ⚠️		softkleenex
softkleenex
💻 ⚠️
mercenarioZ
mercenarioZ
💻 ⚠️		plastictaste
plastictaste
💻 ⚠️ 📖		落叶
落叶
💻 📖 ⚠️ 🌍		n3crosis
n3crosis
💻 ⚠️		copilot
copilot
💻		geoHeil
geoHeil
💻 ⚠️		Jonáš Sivek
Jonáš Sivek
💻 ⚠️
Rubén Pérez Bachiller
Rubén Pérez Bachiller
💻 ⚠️ 🐛		Guanwei Chen
Guanwei Chen
💻 ⚠️		jawwadfirdousi
jawwadfirdousi
💻 ⚠️		Quang Do
Quang Do
💻 ⚠️		Abdullah Alzeiby
Abdullah Alzeiby
💻 ⚠️		zvladru
zvladru
💻 ⚠️		anime girl
anime girl
💻 ⚠️
evaldass
evaldass
💻 ⚠️		Claude
Claude
💻		Shawn
Shawn
💻		luawl
luawl
💻 ⚠️		knightcn1983
knightcn1983
💻 ⚠️		Hafiy
Hafiy
💻 ⚠️ 📖		Kevin van Zonneveld
Kevin van Zonneveld
💻
Choong Jun Jin
Choong Jun Jin
💻 ⚠️		Roman Leventov
Roman Leventov
💻		AliReza Tofighi
AliReza Tofighi
📖		DOMANHDUC
DOMANHDUC
💻		Kwan Perry
Kwan Perry
💻		James
James
💻 ⚠️
- - - - - - -This project follows the [all-contributors](https://github.com/all-contributors/all-contributors) specification. Contributions of any kind welcome! +codex-lb is licensed under the [MIT License](LICENSE). Copyright and attribution remain with the upstream authors and contributors.