TomorrowEdge / 明日边缘

明日边缘是一个 TUI-first multi-model agent cockpit for full-access coding workflows：面向 full-access 代码任务的终端驾驶舱，用来监督、调度、审查、授权多个 AI agents 协作完成软件工程任务。

Full autonomy, full visibility.
完全自治，完全可见。

它不是聊天机器人，也不是某个模型的一层 CLI 壳。Codex / Claude Code gives agents full access. TomorrowEdge gives full access a cockpit.

Full 模式是完整工作区工具权限下的自治执行。TomorrowEdge 会自动应用 patch、运行 shell、执行 repair loop，并继续迭代，不会每一步都打断用户确认。

它与黑盒 full-access agent 的区别不是限制权限，而是可见性：每次模型调用、上下文选择、patch、命令、review、judge 裁决、fallback、成本更新和验证结果都会显示在 TUI 中，并写入可回放事件账本。

为什么存在

明日边缘存在的原因，是 AI coding 的未来不会是单模型的。不同模型有不同的能力、价格、上下文长度、延迟与隐私边界。模型厂有动力把用户留在自家模型栈里，但工程团队真正需要的是跨模型的最优组合：用强模型做高价值判断，用高性价比模型做大规模执行，用本地模型守住隐私，用人类授权关键动作。明日边缘就是这个中立编排层，把异构模型组织成一个可监督、可审计、可回滚的软件工程工作流。 OpenRouter 解决“怎么调用多个模型”；TomorrowEdge 解决“怎么让多个模型在一个真实工程任务里分工、争辩、监督、交付”。

当前版本

当前版本：0.5.0。

这一版的重点是 MCP Agent Bridge：让 Claude Code / Codex 等外部 coding agents 通过 MCP 接入 TomorrowEdge，并被用户绑定到 core、planner、reviewer、judge、coder_a、repairer 等 workflow roles。

tedge mcp serve 启动 TomorrowEdge MCP stdio server
tedge mcp tools 查看暴露给外部 agent 的 MCP tools
tedge mcp agents 查看当前启用的 external MCP agents
external_agents 配置块支持 Claude Code / Codex mock profile 和角色允许列表
agents.<role>.provider: external:<id> 支持把 workflow role 显式绑定到外部 agent
外部 agent 的 patch、review、judgment、result、cost usage 都会写入 events.jsonl
TUI 的 Agents / Router / Trace panes 会显示 external agent badge、role binding 和 external_agent_* 事件
0.5.0 增加 Conversation Targets，并保留 npm run verify、非 git 压缩包 secret scan、full access shell policy、external command runner skeleton，以及本地可运行的 tiny LM demo。

TUI 实际运行截图

以下截图来自当前配置下的 TomorrowEdge fixture 工作流运行态，不是官网页、生成图或概念图。

快速开始

node --version   # requires Node >=20.19.0
npm install
npm test
npm run dev -- doctor
npm run verify
npm run dev -- init
npm run dev -- init --force
npm run dev -- run "fix failing test" --headless
npm run dev -- run "fix failing test" --headless --fixture-mode --approve-patch --approve-shell
npm run dev -- tui

默认测试和演示都可以离线运行，不需要 API key。云端 provider 只有在显式配置环境变量后才会启用；启用后 tedge run 会优先尝试非破坏性 live 候选，必要时仍可用 --offline 回到纯离线 fixture/mock 路径。 npm run dev 在 WSL 且临时目录落到 Windows mount 时会自动把 TMPDIR 切到 /tmp，避免 tsx IPC socket 失败。

核心能力

角色化 agent 图：Planner、Explorer、Coder-A/B、Reviewer、Judge、Runner、Repairer、Summarizer
多模型路由：OpenRouter、DeepSeek、MiMo、Ollama、本地 mock/fixture、OpenAI-compatible 等
能力拼接式路由：图片/截图/流程图先交给 Vision Agent，再把结构化规格交给 coding agent
访问模式：restricted、partial、full
非破坏性 live advisory：真实模型给计划/实现/评审/裁决建议，但不改文件
非破坏性 live patch：真实模型生成候选 diff，但不会自动应用
provider fallback：主路由不可用时按计划 fallback，并在 modelNotes 里记录原因
多模型 drill：Core/River 作为 planner/reviewer，比较不同模型完成同一任务的能力
Core-led workflow：任务拆分、多轮模型辩论、角色执行、Core 审核、报告落盘
MCP Agent Bridge：把 Claude Code / Codex 等外部 coding agents 绑定为 core/planner/reviewer/judge/coder/repairer
patch 安全：预览、敏感文件拦截、显式授权、undo snapshot
产品化安全基线：shell guard、artifact 脱敏、crypto ID、patch 回滚、任务相关上下文选择
TUI 驾驶舱：agent 状态、路由、辩论、diff、shell、证据、记忆、帮助面板

常用命令

tedge init
tedge tui
tedge tui --to reviewer
tedge targets
tedge ask --to reviewer "is this patch safe?"
tedge run "task"
tedge run --to debate "task"
tedge run "task" --headless
tedge run "task" --live
tedge run "task" --offline
tedge config
tedge models
tedge models --real-smoke
tedge models --smoke-suite
tedge mode restricted
tedge mode partial
tedge mode full
tedge prefs
tedge drill "task"
tedge workflow "task"
tedge mcp serve
tedge mcp tools
tedge mcp agents
tedge replay latest
tedge trace latest
tedge trace latest --verbose
tedge export latest --format markdown
tedge export latest --brief
tedge export latest --format json --include-artifacts
tedge sessions
tedge memory
tedge review-export latest --format github
tedge undo --list
tedge undo

权限模式

tedge mode restricted
tedge mode partial
tedge mode full
tedge run "task" --access-mode restricted

restricted：禁止云模型调用和本地变更
partial：允许模型调用，但 patch/shell/repair 需要授权
full：自治执行；自动应用 patch、运行 shell、执行 repair loop，并把每一步写入事件账本

full 会自动批准 patch/shell/repair。CLI 会在进入 full autonomy 时输出风险提示；建议先在 clean repo、sandbox 或 fixture 中使用。

Fixture 演示

完整 approved patch/test loop：

tedge run "fix failing test" --headless --fixture-mode --approve-patch --approve-shell

失败测试后的 Repairer loop：

tedge run "fix failing test" --headless --fixture-mode --approve-patch --approve-shell --fixture-failing-patch --repair-on-fail --approve-repair

没有 --approve-patch 不会应用 diff；没有 --approve-shell 不会运行测试；没有 --approve-repair 只会记录 repair candidate。从 TomorrowEdge 项目根目录运行 fixture demo 时，CLI 会复制 tests/fixtures/sample-repo-basic 到临时目录执行；headless 输出中的 fixtureWorkspace 会显示实际执行目录。

多模型工作流

非破坏性能力 drill：

tedge drill "fix the failing add test" --fixture sample-repo-basic --providers openrouter,deepseek,mimo
tedge drill "restore the login screen from the screenshot" --fixture sample-repo-react-ui --providers openrouter,deepseek,mimo

完整 Core-led workflow：

tedge workflow "design and land a real multi-model orchestration workflow" --providers openrouter,deepseek,mimo
tedge workflow "design and land a real multi-model orchestration workflow" --providers openrouter,deepseek,mimo --rounds 2

workflow 支持 1-5 轮辩论。第 1 轮是角色发言，后续轮次是交叉质询：模型会围绕上轮 transcript 里的矛盾、授权边界和落地风险互相挑战。每个 live batch 都会按 debate.max_cost_usd 做预算预检。

MCP Agent Bridge

TomorrowEdge 不替代 Claude Code / Codex，而是把它们纳入 full-access multi-model cockpit。Codex / Claude Code gives agents full access. TomorrowEdge gives full access a cockpit. TomorrowEdge 不替代你已经订阅的 Claude Code / Codex，而是把它们变成可编排、可观测的角色节点。

MCP bridge 允许外部 coding agents 承担 core、planner、reviewer、judge、coder_a、repairer 等角色。TomorrowEdge 继续负责 orchestration、routing、trace、event ledger、session export 和 TUI 可视化监督。详见 docs/MCP_AGENT_BRIDGE.md 和 docs/EXTERNAL_AGENT_ROLES.md。

基本用法：

tedge mcp tools
tedge mcp agents
tedge mcp agents --probe
tedge mcp serve
tedge mcp invoke codex --session latest --role reviewer --prompt "review the current workflow"
tedge trace latest --verbose

TomorrowEdge 也不浪费你已经订阅的 Claude Code / Codex：它可以把这些昂贵强 agent 绑定到 planner、reviewer、judge 等关键角色，把大规模探索和实现交给更便宜或本地的模型，从而降低全流程强模型成本。 external_agents.<id>.command / args / cwd / env 可用于 command runner skeleton。外部进程通过 stdin 和 TOMORROWEDGE_EXTERNAL_CONTEXT_FILE 接收结构化任务上下文，stdout/stderr 会作为 artifact 写入 trace。

本地 tiny LM demo

cd examples/tiny-local-lm
npm install
npm start
npm run verify

这个 demo 是本地中英双语 hashed neural n-gram toy language model，默认约 54 万参数，不调用 OpenAI/OpenRouter API。它提供 /health、/model-info、/generate，前端支持 prompt、temperature 和 max tokens，用于验证 TomorrowEdge 的多 agent 分工、review、judge、repair 和 export 流程。

角色绑定示例：

external_agents:
  claude_code:
    enabled: true
    transport: mcp
    roles: [core, planner, reviewer, judge]
    capabilities: [core, planning, review, judgment]
    trustLevel: high
  codex:
    enabled: true
    transport: mcp
    command: codex
    args: [mcp-server]
    autoStart: true
    roles: [core, coder_a, repairer, reviewer]
    capabilities: [core, coding, repair, review]
    trustLevel: high

agents:
  planner:
    provider: external:claude_code
    model: auto
  reviewer:
    provider: external:codex
    model: auto
  judge:
    provider: external:claude_code
    model: auto

能力拼接

tedge run "根据截图还原 React 页面" --image ./screen.png --headless

当任务包含图片输入时，TomorrowEdge 会自动插入 Vision Agent：

Image / Screenshot / Diagram
  -> Vision Agent
  -> Structured Visual Spec
  -> Planner / Coder
  -> Patch / Test
  -> Reviewer / Runner

这就是能力拼接式模型路由：不是选择一个模型完成所有事情，而是组合一组最合适的能力。OpenRouter 路由请求，TomorrowEdge 路由能力。详见 docs/CAPABILITY_STITCHING.md。

安全边界

默认 safe mode
patch 和 shell 默认都需要显式授权
ignored/sensitive 文件不会进入上下文选择
suspected secrets 上传云模型前会被拦截
shell 命令不再通过 shell: true 执行；危险命令和 shell 元字符会被拦截
事件 artifact 默认脱敏后再保存和导出
多文件 patch 写入失败时会回滚已写入文件
telemetry 默认关闭
.env 和 .tomorrowedge/ 本地运行态被 git 忽略
provider fallback 会显式记录，不会伪装成主 provider 成功

Provider

Provider	Adapter type	Default enabled	Live smoke	Vision	Status
`mock` / `fixture`	built-in offline	yes	n/a	fixture	stable
OpenRouter	OpenAI-compatible	no	yes, with key	model-dependent	usable
DeepSeek	OpenAI-compatible	no	yes, with key	no/limited	usable
MiMo	OpenAI-compatible	no	yes, with key	supported when model supports images	usable
OpenAI-compatible	generic compatible endpoint	no	yes, with key/base URL	model-dependent	usable
Kimi	Moonshot OpenAI-compatible (`kimi-k2.6`)	no	yes, with key	model-dependent	usable
Ollama	local	yes	local daemon	model-dependent	usable/local
Anthropic	placeholder	no	no	no	planned/native adapter not implemented
Gemini	placeholder	no	no	no	planned/native adapter not implemented

Anthropic/Gemini 目前是显式 placeholder；如果要用 Claude/Opus 或 Gemini 类模型，推荐先通过 OpenRouter 路由，直到 native adapter 实现。

本项目不是 Xiaomi、MiMo、OpenAI、Anthropic、Google、DeepSeek、Moonshot/Kimi 或 OpenRouter 的官方项目。

UI 风格

明日边缘默认中文 UI，整体风格是简约、克制、偏程序员审美的 terminal cockpit：深色面板、细边框、等宽字体、高信息密度、有限状态色，以及不抢主次的轻科幻工程感。详见 docs/UI_STYLE.md。

Clean Room

见 docs/CLEAN_ROOM_NOTE.md。

English

TomorrowEdge is a TUI-first multi-model agent cockpit for coding tasks: a terminal cockpit for supervising, routing, reviewing, authorizing, and auditing multiple AI agents working together on real software engineering workflows. TomorrowEdge is a TUI-first multi-model agent cockpit for full-access coding workflows. TomorrowEdge is not another agent framework. It is a full-access cockpit for native workflows and existing agent frameworks.

Full autonomy, full visibility.

It is not a chatbot CLI and not a single-provider wrapper. Codex / Claude Code gives agents full access. TomorrowEdge gives full access a cockpit.

Full mode is autonomous execution with complete workspace tool access. TomorrowEdge will apply patches, run shell commands, execute repair loops, and continue iterating without per-step confirmation.

The difference from black-box full-access agents is visibility: every model call, context selection, patch, command, review, judge decision, fallback, cost update, and verification result is rendered in the TUI and saved to a replayable event ledger.

Why It Exists

TomorrowEdge exists because the future of AI coding will not be single-model. Different models have different capabilities, prices, context lengths, latency profiles, and privacy boundaries. Model vendors have incentives to keep users inside their own stacks, but engineering teams need the best cross-model composition: use strong models for high-value judgment, cost-efficient models for large-scale execution, local models for privacy, and humans for critical authorization. TomorrowEdge is that neutral orchestration layer. It organizes heterogeneous models into a supervised, auditable, reversible software engineering workflow. OpenRouter solves "how to call multiple models"; TomorrowEdge solves "how to make multiple models divide work, debate, supervise, and deliver inside a real engineering task."

Current Version

Current version: 0.5.0.

This release introduces Conversation Targets: the user can choose who a natural-language message is addressed to while TomorrowEdge still owns orchestration, trace, event ledger, and full-access supervision.

tedge targets lists available targets such as core, planner, reviewer, judge, debate, and enabled agent:<id> external agents
tedge ask --to reviewer "..." records a non-mutating directed conversation trace
tedge run --to debate "..." runs a full workflow while recording the user's chosen conversation target in events.jsonl, TUI, trace, and export
the TUI Goal pane shows Talk to: <target> so the cockpit makes the current communication object explicit
conversation_target and conversation_message events make the handoff replayable and auditable
reviewer/judge quality gates are stricter: parseable diffs, target matching, verification plan, encoding hygiene, and blocking concerns affect automatic selection
context selection no longer treats common binary/image assets as safe text context
the local LM demo now reports roughly 50M parameters by default

The previous MCP Agent Bridge remains available: Claude Code / Codex and other external coding agents can connect through MCP and be bound to workflow roles such as core, planner, reviewer, judge, coder_a, and repairer.

tedge mcp serve starts the TomorrowEdge MCP stdio server
tedge mcp tools lists the MCP tools exposed to external agents
tedge mcp agents lists currently enabled external MCP agents
external_agents config supports Claude Code / Codex mock profiles and role allowlists
agents.<role>.provider: external:<id> binds a workflow role to an external agent
external patch, review, judgment, result, and cost usage submissions are written to events.jsonl
the TUI Agents / Router / Trace panes show external agent badges, role bindings, and external_agent_* events
0.5.0 adds Conversation Targets while preserving the hardened release lane: npm run verify, zip-safe secret scanning, full-access shell policy, an external command runner skeleton, and a locally runnable tiny LM demo.

TUI Runtime Screenshots

These screenshots are captured from the current TomorrowEdge fixture runtime with the current provider configuration. They are not docs-site screenshots or generated concept art.

Quickstart

node --version   # requires Node >=20.19.0
npm install
npm test
npm run dev -- doctor
npm run verify
npm run dev -- init
npm run dev -- init --force
npm run dev -- run "fix failing test" --headless
npm run dev -- run "fix failing test" --headless --fixture-mode --approve-patch --approve-shell
npm run dev -- tui

All default tests and demos run offline without API keys. Cloud providers are disabled unless explicitly configured with environment variables; once enabled, tedge run prefers non-mutating live candidates, and --offline returns to the pure fixture/mock path. When the fixture demo is launched from the TomorrowEdge project root, the CLI copies tests/fixtures/sample-repo-basic into a temporary workspace; headless output reports the actual path as fixtureWorkspace. On WSL, npm run dev automatically switches TMPDIR to /tmp when the inherited temp directory points at a Windows mount, avoiding tsx IPC socket failures.

Core Features

Role-conditioned agent graph: Planner, Explorer, Coder-A/B, Reviewer, Judge, Runner, Repairer, Summarizer
Multi-model routing across OpenRouter, DeepSeek, MiMo, Ollama, local mock/fixture, OpenAI-compatible providers, and placeholders
User-configurable provider/model assignment per agent role for controlled model-comparison experiments
Capability stitching: image/screenshot/diagram inputs go through Vision Agent before coding agents
Access modes: restricted, partial, full
First-class event ledger with replayable events.jsonl and per-session artifacts
Artifact-aware trace/export for diffs, reviews, judge decisions, stdout/stderr, and model call refs
Non-mutating live advisory from routed providers
Non-mutating live patch candidates from routed providers
Explicit provider fallback recorded in modelNotes
Multi-model capability drills with Core/River as planner/reviewer
Core-led workflow with decomposition, multi-round debate, role execution, Core review, and saved reports
MCP Agent Bridge for binding Claude Code / Codex or other external coding agents to core/planner/reviewer/judge/coder/repairer roles
Patch safety: preview, sensitive-file blocking, explicit approval, undo snapshots
Productized safety baseline: guarded shell execution, artifact redaction, crypto IDs, patch rollback, and task-relevant context selection
TUI cockpit panes for agents, routing, debate, diffs, shell, evidence, memory, and help
Conversation Targets for core, role-specific questions, debate-room broadcasts, and external agents
Framework-agnostic orchestration backend abstraction with native as the default backend and LangGraph/CrewAI/AutoGen placeholders

Commands

tedge init
tedge tui
tedge tui --to reviewer
tedge targets
tedge ask --to reviewer "is this patch safe?"
tedge run "task"
tedge run --to debate "task"
tedge run "task" --headless
tedge run "task" --live
tedge run "task" --offline
tedge config
tedge models
tedge models --real-smoke
tedge mode restricted
tedge mode partial
tedge mode full
tedge prefs
tedge drill "task"
tedge workflow "task"
tedge mcp serve
tedge mcp tools
tedge mcp agents
tedge replay latest
tedge trace latest
tedge trace latest --verbose
tedge export latest --format markdown
tedge export latest --brief
tedge export latest --format json --include-artifacts
tedge sessions
tedge undo --list
tedge undo

Conversation Targets

TomorrowEdge Core is the default natural-language conversation object. Users can also address a specific role or external agent while the cockpit keeps ownership of orchestration, routing, trace, session export, and supervision.

tedge targets
tedge ask --to core "what should happen next?"
tedge ask --to reviewer "is this diff safe to approve?"
tedge ask --to judge "should we select or request revision?"
tedge ask --to agent:codex "review the latest session"
tedge run --to debate "implement this feature after multi-agent debate"

Every directed message records conversation_target and conversation_message events. Markdown and JSON exports include the chosen target, and the TUI Goal pane shows Talk to: <target>.

Access Modes

restricted: blocks cloud/model calls and local mutations
partial: allows model calls while requiring patch/shell/repair approval
full: autonomous execution with complete workspace tool access; patch/shell/repair loop actions are auto-approved and logged

full auto-approves patch, shell, and repair actions. The CLI prints a risk warning before full-autonomy runs; prefer a clean repo, sandbox, or fixture.

Workflow

tedge drill "fix the failing add test" --fixture sample-repo-basic --providers openrouter,deepseek,mimo
tedge workflow "design and land a real multi-model orchestration workflow" --providers openrouter,deepseek,mimo --rounds 2

workflow supports 1-5 debate rounds. Later rounds are cross-examination rounds over the prior transcript. Each live batch is preflighted against debate.max_cost_usd.

MCP Agent Bridge

TomorrowEdge is not replacing Claude Code / Codex. It turns them into role-bound agents inside a visible multi-model cockpit. Codex / Claude Code gives agents full access. TomorrowEdge gives full access a cockpit.

The MCP bridge lets external coding agents take roles such as core, planner, reviewer, judge, coder_a, and repairer. TomorrowEdge keeps orchestration, routing, trace, event ledger, session export, and TUI supervision. See docs/MCP_AGENT_BRIDGE.md and docs/EXTERNAL_AGENT_ROLES.md. TomorrowEdge does not replace the Claude Code / Codex subscriptions you already have. It turns them into orchestratable and observable role nodes.

Basic usage:

tedge mcp tools
tedge mcp agents
tedge mcp agents --probe
tedge mcp serve
tedge mcp invoke codex --session latest --role reviewer --prompt "review the current workflow"
tedge trace latest --verbose

It also protects existing Claude Code / Codex subscriptions by binding expensive strong agents to high-value roles such as planner, reviewer, and judge while cheaper or local models handle broad execution. external_agents.<id>.command / args / cwd / env can also drive the command runner skeleton. The process receives structured task context through stdin and TOMORROWEDGE_EXTERNAL_CONTEXT_FILE; stdout/stderr are stored as trace artifacts.

Local Tiny LM Demo

cd examples/tiny-local-lm
npm install
npm start
npm run verify

The demo is a local bilingual Chinese/English hashed neural n-gram toy language model with roughly 50M parameters by default, not an OpenAI or OpenRouter API call. It exposes /health, /model-info, and /generate, plus a frontend with prompt, temperature, and max token controls for orchestration acceptance drills.

Role binding example:

external_agents:
  claude_code:
    enabled: true
    transport: mcp
    roles: [core, planner, reviewer, judge]
    capabilities: [core, planning, review, judgment]
    trustLevel: high
  codex:
    enabled: true
    transport: mcp
    command: codex
    args: [mcp-server]
    autoStart: true
    roles: [core, coder_a, repairer, reviewer]
    capabilities: [core, coding, repair, review]
    trustLevel: high

agents:
  planner:
    provider: external:claude_code
    model: auto
  reviewer:
    provider: external:codex
    model: auto
  judge:
    provider: external:claude_code
    model: auto

Orchestration Backends

TomorrowEdge keeps the cockpit contract even when execution is delegated:

orchestration:
  backend: native # native | langgraph | crewai | autogen

native is executable today and wraps the current TomorrowEdge agent graph. langgraph, crewai, and autogen are registered placeholders with schema, docs, and clear unavailable-backend errors. External frameworks are adapters; they do not own full-access authorization, the event ledger, replay, export, or TUI visibility.

See docs/ORCHESTRATION_BACKENDS.md.

Capability Stitching

tedge run "restore this React page from the screenshot" --image ./screen.png --headless

When image input is present, TomorrowEdge inserts a Vision Agent:

Image / Screenshot / Diagram
  -> Vision Agent
  -> Structured Visual Spec
  -> Planner / Coder
  -> Patch / Test
  -> Reviewer / Runner

This is capability compositional routing: do not choose one model to do everything; compose the right capability chain for the task. OpenRouter routes requests. TomorrowEdge routes capabilities. See docs/CAPABILITY_STITCHING.md.

Safety

Safe mode is enabled by default
Patch and shell actions require approval by default
Ignored and sensitive files are excluded from context selection
Suspected secrets are blocked before cloud upload
Shell commands run without shell: true; metacharacters and dangerous executables are blocked
Event artifacts are redacted before persistence/export
Multi-file patch writes roll back if a later write fails
Telemetry is disabled by default
.env and local .tomorrowedge/ runtime state are git-ignored
Provider fallback is explicit; it does not hide the failed primary route

Providers

Provider	Adapter type	Default enabled	Live smoke	Vision	Status
`mock` / `fixture`	built-in offline	yes	n/a	fixture	stable
OpenRouter	OpenAI-compatible	no	yes, with key	model-dependent	usable
DeepSeek	OpenAI-compatible	no	yes, with key	no/limited	usable
MiMo	OpenAI-compatible	no	yes, with key	supported when model supports images	usable
OpenAI-compatible	generic compatible endpoint	no	yes, with key/base URL	model-dependent	usable
Kimi	Moonshot OpenAI-compatible (`kimi-k2.6`)	no	yes, with key	model-dependent	usable
Ollama	local	yes	local daemon	model-dependent	usable/local
Anthropic	placeholder	no	no	no	planned/native adapter not implemented
Gemini	placeholder	no	no	no	planned/native adapter not implemented

Anthropic/Gemini are explicit placeholders today. Route Claude/Opus or Gemini-class models through OpenRouter until native adapters are implemented.

Recommended bilingual config / 推荐配置:

providers:
  openrouter:
    enabled: true
    api_key_env: OPENROUTER_API_KEY
    base_url: https://openrouter.ai/api/v1
    model: openai/gpt-5.2
    api_format: openai_chat
    auth_header: bearer
  deepseek:
    enabled: true
    api_key_env: DEEPSEEK_API_KEY
    base_url: https://api.deepseek.com
    model: deepseek-v4-pro
    api_format: openai_chat
    auth_header: bearer
  mimo:
    enabled: true
    api_key_env: MIMO_API_KEY
    base_url: https://token-plan-sgp.xiaomimimo.com/v1
    model: mimo-v2.5-pro
    api_format: openai_chat
    auth_header: api-key

agents:
  vision: { provider: mimo, model: mimo-v2.5-pro }
  planner: { provider: openrouter, model: openai/gpt-5.2 }
  explorer: { provider: deepseek, model: deepseek-v4-pro }
  coder_a: { provider: deepseek, model: deepseek-v4-pro }
  reviewer: { provider: openrouter, model: anthropic/claude-opus-4.1 }
  judge: { provider: openrouter, model: openai/gpt-5.2 }

This is a recommended starting point, not a hardcoded assignment. Users can replace providers.<id>.model or any agents.<role>.provider/model entry to compare GPT, Claude/Opus, DeepSeek, MiMo, Kimi, Ollama, or any compatible model. auth_header supports bearer, api-key, and none; api_format supports openai_chat and legacy_chat.

This is not an official Xiaomi, MiMo, OpenAI, Anthropic, Google, DeepSeek, Moonshot/Kimi, or OpenRouter project.

UI Style

TomorrowEdge defaults to Chinese UI copy and uses a restrained programmer-facing terminal cockpit style: dark panes, thin borders, monospaced text, dense state panels, limited status colors, and subtle sci-fi engineering accents that never outrank the workflow. See docs/UI_STYLE.md.

Name		Name	Last commit message	Last commit date
Latest commit History 33 Commits
.github		.github
docs		docs
examples		examples
scripts		scripts
src		src
tests		tests
.env.example		.env.example
.gitignore		.gitignore
CHANGELOG.md		CHANGELOG.md
CODE_OF_CONDUCT.md		CODE_OF_CONDUCT.md
CONTRIBUTING.md		CONTRIBUTING.md
LICENSE		LICENSE
README.md		README.md
package-lock.json		package-lock.json
package.json		package.json
tsconfig.json		tsconfig.json
vitest.config.ts		vitest.config.ts

Folders and files

Latest commit

History

Repository files navigation

TomorrowEdge / 明日边缘

为什么存在

当前版本

TUI 实际运行截图

快速开始

核心能力

常用命令

权限模式

Fixture 演示

多模型工作流

MCP Agent Bridge

本地 tiny LM demo

能力拼接

安全边界

Provider

UI 风格

Clean Room

English

Why It Exists

Current Version

TUI Runtime Screenshots

Quickstart

Core Features

Commands

Conversation Targets

Access Modes

Workflow

MCP Agent Bridge

Local Tiny LM Demo

Orchestration Backends

Capability Stitching

Safety

Providers

UI Style

About

Resources

License

Code of conduct

Contributing

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages