Files
deepseek-harness/docs/user/develop/practice/llm-adapter.zh.md
T
imccyu ec601ca13d build(vendor): rescope the vendored Cordis packages into @deepseek-ai
Machine-produced by `pnpm run rescope-vendor --apply` plus the regeneration it
prints: `pnpm install` for the lockfile, `pnpm run gen-third-party-notices`,
`verify-translation-pairing --write` for the touched bilingual pairs,
`gen-doc-graphs`, and one typert snapshot whose ids embed character offsets.
`pnpm run rescope-vendor --check` verifies the result.

Renames nine vendored packages (cordis, cosmokit, schemastery and the six
@cordisjs plugins) and every reference that resolves them: manifest names and
dependency keys, module specifiers including declare-module merges, cordis.yml
plugin names, tsconfig paths, every Markdown fence, and `docs/` prose.
Directory names, upstream versions, and dependency ranges are unchanged, so
vendor/README.md still reads as an upstream snapshot; its manifest table gains
an upstream-name column so THIRD_PARTY_NOTICES keeps MIT attribution pointed
at each fork's origin.

The tutorial tier follows the rename end to end: its yaml fences named plugins
the Loader can no longer resolve, its `ts ignore-check` fences disagreed with
the compiled fences beside them, and its prose quoted both. The contracts that
told readers to keep upstream names — the root convention and the vendoring
cookbook's tree comment and manifest invariant — now say to rescope instead.

Two rules read `@deepseek-ai/` as "another workspace plugin": the client bundle
purity gate now names the vendored libraries a browser bundle inlines, and the
files where a bare `cordis` is an agent-preset id keep that product data.
2026-08-10 22:04:13 +08:00

191 lines
6.2 KiB
Markdown

# LLM 适配器
[English](llm-adapter.md) | 中文
本文介绍如何为 Harness 接入新的模型提供方。
## 概述
LLM 适配器是一个继承 `LlmAdapter` 并实现 `stream()` 方法的类,它会将 Harness 的提供方无关请求转换为具体提供方的 API 调用,并将响应转换回 Harness 分片。
## 最小实现
```ts
import type { Context } from '@deepseek-ai/cordis'
import Schema from '@deepseek-ai/schemastery'
import { LlmAdapter, type GenerateOptions, type StreamChunk } from '@deepseek-ai/dsh-llm'
class MyAdapter extends LlmAdapter {
private apiKey: string
constructor(apiKey: string) {
super()
this.apiKey = apiKey
}
async *stream(options: GenerateOptions): AsyncIterable<StreamChunk> {
// 1. Convert options.messages to the provider format.
// 2. Call the streaming API.
// 3. Convert the response into StreamChunk values.
}
}
export interface Config {
apiKey: string
models: string[]
}
export const Config: Schema<Config> = Schema.object({
apiKey: Schema.string().required(),
models: Schema.array(Schema.string()).required(),
})
export const name = 'my-llm-adapter'
export const inject = ['llm']
export function apply(ctx: Context, config: Config) {
const adapter = new MyAdapter(config.apiKey)
ctx.llm.registerAdapter(config.models, adapter)
}
```
## StreamChunk 协议
`stream()` 必须按以下协议生成分片:
```ts
import { CallId, type StreamChunk } from '@deepseek-ai/dsh-llm'
async function* exampleChunks(): AsyncIterable<StreamChunk> {
// 1. Start each content block with block-start.
yield { type: 'block-start', index: 0, blockType: 'text' }
// 2. Stream text through text-delta.
yield { type: 'text-delta', index: 0, text: 'Hello' }
yield { type: 'text-delta', index: 0, text: ' world' }
// 3. End each content block with block-end and the complete block.
yield {
type: 'block-end',
index: 0,
block: { type: 'text', text: 'Hello world' },
}
// 4. Tool-call block.
yield { type: 'block-start', index: 1, blockType: 'tool-call' }
yield {
type: 'tool-call-delta',
index: 1,
id: CallId('call-123'),
name: 'bash',
argumentsDelta: '{"command":"ls"}',
}
yield {
type: 'block-end',
index: 1,
block: {
type: 'tool-call',
id: CallId('call-123'),
name: 'bash',
arguments: '{"command":"ls"}',
},
}
// 5. Token usage.
yield { type: 'usage', usage: { inputTokens: 100, outputTokens: 50 } }
// 6. Finish reason.
yield { type: 'finish', reason: { kind: 'stop' } }
// Alternatively, { kind: 'tool-calls' } requests tool execution.
}
```
### 关键规则
- 每个 `block-start` 都必须有与之对应的 `block-end`
- `index` 从 0 开始递增,用于标识内容块的顺序。
- `tool-call-delta``argumentsDelta` 是原始 JSON 文本的增量,可以在一个分片中完整生成,也可以分多个分片生成。
- `finish` 必须是最后一个分片。
- `usage` 必须在 `finish` 之前生成。
## GenerateOptions
`stream()` 接收仓库导出的 `GenerateOptions`。它包含模型、适配器拥有的推理强度 ID、对话历史、系统提示词、工具 schema、生成参数、停止序列和中止信号;完整字段以 `@deepseek-ai/dsh-llm` 导出的 TypeScript 类型为准。适配器必须将支持的字段映射到具体 API;如果无法支持某个字段,应抛出带稳定 code 的 `LlmError`,不得静默丢弃。
请覆写 `resolveModel(provider, model, signal?)`,在一次查询中返回确切的提供方/模型身份以及可选的 `context``reasoning` 元数据。推理元数据包含有序的不透明 ID、展示名称,以及可选的配置默认值;请保留适配器给出的权威可选列表,包括其上游能力 API 返回的 `off`,不要将这些值提升为核心枚举。异步查询必须响应该可选信号,使取消和资源释放过程完全停稳。服务会校验聚合结果,并在调用 `stream()` 前拒绝显式指定但不受支持的推理强度;省略 `reasoning` 表示该模型没有可选的推理强度能力。
## 注册适配器
```ts ignore-check
ctx.llm.registerAdapter(['model-name-1', 'model-name-2'], adapter)
```
第一个参数是该适配器支持的模型名列表。当用户在 `cordis.yml` 中配置 `model: model-name-1` 时,框架会将请求路由到该适配器。
## 在 cordis.yml 中使用
```yaml
- id: my-llm
name: './src/my-llm-adapter.ts'
config:
apiKey: !!js process.env.MY_API_KEY
models:
- my-model-v1
- my-model-v2
- id: agent-loop
name: '@deepseek-ai/dsh-agent-loop'
config:
agents:
- id: main
provider: my-llm
model: my-model-v1 # References the model registered above.
workspaceContext: false
```
## 实战参考
仓库中包含以下两个完整实现:
- `packages/llm/llm-deepseek/` — DeepSeek API 适配器(OpenAI 兼容格式)
- `packages/llm/llm-pi-ai/` — Pi AI 适配器(不同的 API 格式)
对比这两个已交付的适配器,可以看到同一套 harness 约定如何在不同提供方 SDK 之上实现。
## 错误处理
适配器应通过带稳定 code 的 `LlmError` 抛出传输和协议故障;agent loop(智能体循环)会保留该错误及其 code,用于诊断和策略处理。不要依赖普通 `Error` 被自动转换。每个提供方 HTTP 请求还必须合并 `attributionHeaders()`,并传递 `options.signal`。
```ts
import {
attributionHeaders,
LlmAdapter,
LlmError,
type GenerateOptions,
type StreamChunk,
} from '@deepseek-ai/dsh-llm'
class HttpAdapter extends LlmAdapter {
constructor(private readonly endpoint: string) {
super()
}
async *stream(options: GenerateOptions): AsyncIterable<StreamChunk> {
const response = await fetch(this.endpoint, {
method: 'POST',
headers: {
'content-type': 'application/json',
...attributionHeaders(),
},
body: JSON.stringify({ model: options.model, messages: options.messages }),
...options.signal ? { signal: options.signal } : {},
})
if (!response.ok) {
throw new LlmError(`Provider API error: ${response.status}`, 'PROVIDER_HTTP_ERROR')
}
// A real adapter parses the response and emits the complete chunk sequence.
yield { type: 'finish', reason: { kind: 'stop' } }
}
}
```