Add abstract service interface packages

@deepseek-ai/dsh-llm: provider-neutral content-block vocabulary
(merge-extensible maps), raw StreamChunk protocol, ToolSchema,
abstract LlmAdapter, LlmService adapter registry, BlockAssembler.

@deepseek-ai/dsh-session: event-sourced Session (append-only log,
deriveMessages; context/steering render as tagged envelopes),
SessionStore, session/event + awaited session/flush durability seam.

@deepseek-ai/dsh-system-prompt: ordered sections + tool-schema
providers; assemble() through the system-prompt/assemble waterfall.
Tool schemas are part of the assembly by design.

@deepseek-ai/dsh-tools: tool registry feeding schemas into the
assembly; execute() through the tools/execute waterfall (the single
sandbox/permission/hook seam).

@deepseek-ai/dsh-agent: Agent interface (send/steer/inject/abort,
spawn/fork TODO seams), AgentRegistry, and the full agent/* event
taxonomy so plugins never depend on the concrete loop.
This commit is contained in:
Tianyi Cui
2026-06-11 10:54:06 +08:00
parent 72688a3888
commit d5a1d9bb75
25 changed files with 1703 additions and 0 deletions
+32
View File
@@ -0,0 +1,32 @@
{
"name": "@deepseek-ai/dsh-agent",
"description": "Agent interface, registry, and event vocabulary for the DeepSeek Harness",
"version": "0.0.1",
"private": true,
"type": "module",
"main": "lib/index.js",
"types": "lib/index.d.ts",
"exports": {
".": {
"types": "./lib/index.d.ts",
"default": "./lib/index.js"
},
"./src/*": "./src/*",
"./package.json": "./package.json"
},
"files": [
"lib",
"src"
],
"license": "BSD-3-Clause",
"peerDependencies": {
"@deepseek-ai/dsh-llm": "^0.0.1",
"@deepseek-ai/dsh-session": "^0.0.1",
"cordis": "^4.0.0-rc.6"
},
"devDependencies": {
"@deepseek-ai/dsh-llm": "^0.0.1",
"@deepseek-ai/dsh-session": "^0.0.1",
"cordis": "^4.0.0-rc.6"
}
}
+49
View File
@@ -0,0 +1,49 @@
import { Context, Service } from 'cordis'
import type { Agent } from './types.ts'
export * from './types.ts'
declare module 'cordis' {
interface Context {
agents: AgentRegistry
}
}
/**
* Agent registry (`ctx.agents`): tracks live agents so UI, hook, and
* orchestrator plugins can find them without depending on the concrete loop
* package. Agent *creation* belongs to whichever plugin implements the Agent
* interface (phase 1: `@deepseek-ai/dsh-agent-loop`).
*/
export class AgentRegistry extends Service {
private store = new Map<string, Agent>()
constructor(ctx: Context) {
super(ctx, 'agents')
}
/** Register a live agent. Disposed with the calling fiber. */
register(agent: Agent): () => void {
return this.ctx.effect(() => {
if (this.store.has(agent.id)) {
throw new Error(`agent "${agent.id}" is already registered`)
}
this.store.set(agent.id, agent)
this.ctx.emit('agent/created', agent)
return () => {
this.store.delete(agent.id)
this.ctx.emit('agent/disposed', agent)
}
}, 'agents.register()')
}
get(id: string): Agent | undefined {
return this.store.get(id)
}
list(): Agent[] {
return [...this.store.values()]
}
}
export default AgentRegistry
+107
View File
@@ -0,0 +1,107 @@
import type { ContentBlock, GenerateOptions, Message, MessageSource, StreamChunk } from '@deepseek-ai/dsh-llm'
import type { Session, TurnEndReason } from '@deepseek-ai/dsh-session'
/**
* Options an agent is created with.
* Merge-extensible: plugins declare extra fields via declaration merging.
*/
export interface AgentOptions {
/** Model name (must have a registered adapter at call time). */
model?: string
/** Per-agent system prompt appended after the assembled sections. */
systemPrompt?: string
}
export interface SendOptions {
source?: MessageSource
}
export type AgentStatus = 'idle' | 'running' | 'disposed'
/**
* The agent handle — the surface every plugin (UI, hooks, orchestrators)
* programs against. The concrete implementation lives in
* `@deepseek-ai/dsh-agent-loop` (class `LoopAgent`); nothing outside the loop
* package should depend on the implementation.
*/
export interface Agent {
readonly id: string
readonly options: AgentOptions
readonly session: Session
readonly status: AgentStatus
/** Queue a user message. Starts a turn when idle; otherwise waits for the next turn. */
send(content: ContentBlock[], options?: SendOptions): void
/**
* Steer a running turn: content is injected between steps of the current
* turn. When idle, behaves like {@link send}.
*/
steer(content: ContentBlock[], options?: SendOptions): void
/**
* Inject in-session context (file-change notices, skill content, cron
* notifications, …): appends a `context/message` session event without
* triggering a turn — the next model request sees it at its chronological
* position, rendered as tagged synthetic context rather than a user prompt.
*
* TODO(review): exact envelope/rendering rules live in dsh-session and need
* review once a real adapter exists.
*/
inject(content: ContentBlock[], options?: SendOptions): void
/** Abort the in-flight step (if any); the turn ends with reason 'aborted'. */
abort(reason?: string): void
// TODO(sub-agents): spawn/fork seams — semantics deliberately deferred.
// The intended shape: a creation option referencing a parent agent
// (fork = seed the child Session with the parent's event log; spawn =
// fresh Session), with the child returned as an Agent handle so steer()
// and event subscription work uniformly. See docs/architecture.md.
}
declare module 'cordis' {
interface Events {
// ---- lifecycle (emit) ----
/** An agent was registered. */
'agent/created'(agent: Agent): void
/** An agent was disposed. */
'agent/disposed'(agent: Agent): void
/** Agent status changed (idle/running/disposed). */
'agent/status'(agent: Agent, status: AgentStatus): void
/** A message entered the agent's inbox (queued or steering). */
'agent/queued'(agent: Agent, content: ContentBlock[], options: SendOptions & { steering: boolean }): void
// ---- turn/step boundaries (emit) ----
'agent/turn-start'(agent: Agent, turn: number): void
'agent/turn-end'(agent: Agent, turn: number, reason: TurnEndReason): void
'agent/step-start'(agent: Agent, turn: number, step: number): void
'agent/step-end'(agent: Agent, turn: number, step: number): void
// ---- interception seams (waterfall) ----
/**
* Waterfall: mutate the fully-assembled GenerateOptions before the model
* call (hooks, compaction, model switching, tool filtering, …).
*/
'agent/request'(agent: Agent, turn: number, step: number, options: GenerateOptions, next: () => Promise<GenerateOptions>): Promise<GenerateOptions>
/**
* Waterfall: post-process the assembled assistant message before tool
* dispatch (validation, content rewriting, …).
*/
'agent/step-result'(agent: Agent, turn: number, step: number, message: Message, next: () => Promise<Message>): Promise<Message>
/**
* Waterfall: override the turn-continuation decision. The default
* (computed by the loop) is `hadToolCalls || steeringInjected`. Listeners
* can force-continue (/goal, /loop) or force-stop (budget guards).
*/
'agent/turn-continuation'(agent: Agent, turn: number, defaultDecision: boolean, next: () => Promise<boolean>): Promise<boolean>
// ---- streaming + tool notifications (emit) ----
/** A raw stream chunk arrived (token-level UI/log feed). */
'agent/stream-chunk'(agent: Agent, turn: number, step: number, chunk: StreamChunk): void
/** Steering content was injected into a running turn. */
'agent/steering'(agent: Agent, turn: number, content: ContentBlock[]): void
/** A step or turn errored. */
'agent/error'(agent: Agent, turn: number, step: number, error: Error): void
}
}
+54
View File
@@ -0,0 +1,54 @@
import { describe, expect, it } from 'vitest'
import { Context } from 'cordis'
import { Session } from '@deepseek-ai/dsh-session'
import AgentRegistry, { Agent } from '@deepseek-ai/dsh-agent'
function stubAgent(id: string): Agent {
return {
id,
options: {},
session: new Session(`${id}-session`),
status: 'idle',
send() {},
steer() {},
inject() {},
abort() {},
}
}
describe('AgentRegistry', () => {
it('registers agents and emits created/disposed events', async () => {
const ctx = new Context()
await ctx.plugin(AgentRegistry)
const created: string[] = []
const disposed: string[] = []
ctx.on('agent/created', agent => void created.push(agent.id))
ctx.on('agent/disposed', agent => void disposed.push(agent.id))
const agent = stubAgent('a1')
const dispose = ctx.agents.register(agent)
expect(created).toEqual(['a1'])
expect(ctx.agents.get('a1')).toBe(agent)
expect(ctx.agents.list()).toEqual([agent])
dispose()
expect(disposed).toEqual(['a1'])
expect(ctx.agents.get('a1')).toBeUndefined()
})
it('rejects duplicate ids and unregisters on fiber dispose (HMR safety)', async () => {
const ctx = new Context()
await ctx.plugin(AgentRegistry)
ctx.agents.register(stubAgent('main'))
expect(() => ctx.agents.register(stubAgent('main'))).toThrow('already registered')
const fiber = await ctx.plugin(Object.assign((inner: Context) => {
inner.agents.register(stubAgent('scoped'))
}, { inject: ['agents'] }))
expect(ctx.agents.list().map(a => a.id)).toEqual(['main', 'scoped'])
await fiber.dispose()
expect(ctx.agents.list().map(a => a.id)).toEqual(['main'])
})
})
+14
View File
@@ -0,0 +1,14 @@
{
"extends": "../../tsconfig.base.json",
"compilerOptions": {
"rootDir": "src",
"outDir": "lib"
},
"include": ["src"],
"references": [
{ "path": "../../vendor/cosmokit" },
{ "path": "../../vendor/cordis" },
{ "path": "../llm" },
{ "path": "../session" }
]
}
+28
View File
@@ -0,0 +1,28 @@
{
"name": "@deepseek-ai/dsh-llm",
"description": "Provider-neutral LLM service interface for the DeepSeek Harness",
"version": "0.0.1",
"private": true,
"type": "module",
"main": "lib/index.js",
"types": "lib/index.d.ts",
"exports": {
".": {
"types": "./lib/index.d.ts",
"default": "./lib/index.js"
},
"./src/*": "./src/*",
"./package.json": "./package.json"
},
"files": [
"lib",
"src"
],
"license": "BSD-3-Clause",
"peerDependencies": {
"cordis": "^4.0.0-rc.6"
},
"devDependencies": {
"cordis": "^4.0.0-rc.6"
}
}
+115
View File
@@ -0,0 +1,115 @@
import type { ContentBlock, FinishReason, GenerateResult, Message, StreamChunk, TokenUsage } from './types.ts'
/**
* Incrementally assembles raw {@link StreamChunk}s into complete
* {@link ContentBlock}s and a final assistant {@link Message}.
*
* This is the single shared assembly implementation: the agent loop feeds it
* while logging raw chunks for replay fidelity, and `LlmService.generate()` /
* `streamBlocks()` use it to offer assembled views of the same stream.
*/
export class BlockAssembler {
private partials = new Map<number, {
blockType: string
text: string
toolCallId?: string
toolCallName?: string
toolCallArguments: string
block?: ContentBlock
}>()
private order: number[] = []
private _usage: TokenUsage | undefined
private _finish: FinishReason | undefined
/**
* Feed one chunk. Returns the completed block when the chunk closes one
* (either an explicit `block-end` or an implicit close), otherwise undefined.
*/
push(chunk: StreamChunk): ContentBlock | undefined {
switch (chunk.type) {
case 'block-start': {
if (!this.partials.has(chunk.index)) this.order.push(chunk.index)
this.partials.set(chunk.index, {
blockType: chunk.blockType,
text: '',
toolCallArguments: '',
})
return
}
case 'text-delta':
case 'reasoning-delta': {
const partial = this.ensure(chunk.index, chunk.type === 'text-delta' ? 'text' : 'reasoning')
partial.text += chunk.text
return
}
case 'tool-call-delta': {
const partial = this.ensure(chunk.index, 'tool-call')
partial.toolCallId = chunk.id
if (chunk.name) partial.toolCallName = chunk.name
partial.toolCallArguments += chunk.argumentsDelta
return
}
case 'block-end': {
const partial = this.ensure(chunk.index, chunk.block.type)
partial.block = chunk.block
return chunk.block
}
case 'usage': {
this._usage = chunk.usage
return
}
case 'finish': {
this._finish = chunk.reason
return
}
}
}
private ensure(index: number, blockType: string) {
let partial = this.partials.get(index)
if (!partial) {
partial = { blockType, text: '', toolCallArguments: '' }
this.partials.set(index, partial)
this.order.push(index)
}
return partial
}
/** Assemble all blocks seen so far, in stream order. */
blocks(): ContentBlock[] {
return this.order.map((index) => {
const partial = this.partials.get(index)!
if (partial.block) return partial.block
switch (partial.blockType) {
case 'text': return { type: 'text', text: partial.text }
case 'reasoning': return { type: 'reasoning', text: partial.text }
case 'tool-call': return {
type: 'tool-call',
id: partial.toolCallId ?? `call-${index}`,
name: partial.toolCallName ?? '',
arguments: partial.toolCallArguments,
}
default: throw new Error(`cannot assemble incomplete block of type "${partial.blockType}"`)
}
})
}
get usage(): TokenUsage | undefined {
return this._usage
}
get finish(): FinishReason {
return this._finish ?? { kind: 'stop' }
}
/** The assembled assistant message. */
message(): Message {
return { role: 'assistant', content: this.blocks() }
}
/** The assembled non-streaming result. */
result(): GenerateResult {
return { message: this.message(), usage: this._usage, finish: this.finish }
}
}
+108
View File
@@ -0,0 +1,108 @@
import { Context, Service } from 'cordis'
import type { ContentBlock, GenerateOptions, GenerateResult, StreamChunk } from './types.ts'
import { BlockAssembler } from './assembler.ts'
export * from './types.ts'
export { BlockAssembler } from './assembler.ts'
declare module 'cordis' {
interface Context {
llm: LlmService
}
interface Events {
/** Waterfall around every streaming model call (retry, caching, routing). */
'llm/stream'(this: LlmService, options: GenerateOptions, next: () => AsyncIterable<StreamChunk>): AsyncIterable<StreamChunk>
/** Waterfall around every non-streaming model call. */
'llm/generate'(this: LlmService, options: GenerateOptions, next: () => Promise<GenerateResult>): Promise<GenerateResult>
/** An adapter was registered or unregistered. */
'llm/adapter-change'(): void
}
}
export class LlmError extends Error {
constructor(message: string, public code: string) {
super(message)
this.name = 'LlmError'
}
}
/**
* Base class for LLM provider adapters.
*
* An adapter translates between the harness vocabulary (Message/ContentBlock/
* StreamChunk) and one provider's wire format. Adapters register themselves
* via `ctx.llm.registerAdapter(models, adapter)`.
*
* TODO: the first real adapter (DeepSeek V4) lands in a later phase; until
* then only mock adapters (tests, demo) exist.
*/
export abstract class LlmAdapter {
/** Stream one model call as raw chunks. The only required method. */
abstract stream(options: GenerateOptions): AsyncIterable<StreamChunk>
}
/**
* The abstract `llm` service: an adapter registry plus streaming /
* non-streaming call surfaces, both interceptable via waterfall events.
*/
export class LlmService extends Service {
private adapters = new Map<string, LlmAdapter>()
constructor(ctx: Context) {
super(ctx, 'llm')
}
/** Register an adapter for the given model names. Disposed with the fiber. */
registerAdapter(models: string[], adapter: LlmAdapter): () => void {
return this.ctx.effect(() => {
for (const model of models) this.adapters.set(model, adapter)
this.ctx.emit('llm/adapter-change')
return () => {
for (const model of models) this.adapters.delete(model)
this.ctx.emit('llm/adapter-change')
}
}, 'llm.registerAdapter()')
}
/** Model names with a registered adapter. */
models(): string[] {
return [...this.adapters.keys()]
}
private adapter(model: string): LlmAdapter {
const adapter = this.adapters.get(model)
if (!adapter) throw new LlmError(`no adapter registered for model "${model}"`, 'NO_ADAPTER')
return adapter
}
/** Stream one model call as raw chunks (token-level deltas). */
stream(options: GenerateOptions): AsyncIterable<StreamChunk> {
return this.ctx.waterfall(this, 'llm/stream', options, () => {
return this.adapter(options.model).stream(options)
})
}
/**
* Stream one model call as completed content blocks — a convenience view
* for consumers that don't care about token-level deltas.
*/
async * streamBlocks(options: GenerateOptions): AsyncIterable<ContentBlock> {
const assembler = new BlockAssembler()
for await (const chunk of this.stream(options)) {
const block = assembler.push(chunk)
if (block) yield block
}
}
/** One model call, fully assembled (drains the chunk stream). */
generate(options: GenerateOptions): Promise<GenerateResult> {
return this.ctx.waterfall(this, 'llm/generate', options, async () => {
const assembler = new BlockAssembler()
for await (const chunk of this.stream(options)) assembler.push(chunk)
return assembler.result()
})
}
}
export default LlmService
+178
View File
@@ -0,0 +1,178 @@
/**
* Provider-neutral message and streaming vocabulary.
*
* This is the canonical language spoken by the agent loop, session logs, and
* every plugin. Adapters translate it to provider wire formats (DeepSeek V4
* first); nothing outside an adapter should ever see a provider-specific
* shape.
*
* Extensibility: the unions in this file are derived from interfaces
* (`ContentBlockMap`, `MessageSourceMap`, `FinishReasonMap`) so that plugins
* can extend them via declaration merging:
*
* ```ts
* declare module '@deepseek-ai/dsh-llm' {
* interface ContentBlockMap {
* video: { type: 'video'; url: string }
* }
* }
* ```
*/
/** Cache hint attached to a content block (provider-interpreted). */
export type CacheHint = 'ephemeral'
/** Plain text visible to the end user. */
export interface TextBlock {
type: 'text'
text: string
cache?: CacheHint
}
/** Reasoning / thinking content, distinct from visible text. */
export interface ReasoningBlock {
type: 'reasoning'
text: string
}
/** A tool invocation requested by the model. */
export interface ToolCallBlock {
type: 'tool-call'
/** Provider-issued call id; correlates with the matching tool result. */
id: string
name: string
/** Raw JSON string as produced by the model. */
arguments: string
}
/** The result of a tool invocation, sent back to the model. */
export interface ToolResultBlock {
type: 'tool-result'
toolCallId: string
content: ContentBlock[]
isError?: boolean
cache?: CacheHint
}
/** An image, by URL or data URL. */
export interface ImageBlock {
type: 'image'
url: string
mimeType?: string
cache?: CacheHint
}
/**
* All known content block shapes, keyed by their `type` tag.
* Merge-extensible: plugins add new block types via declaration merging.
*/
export interface ContentBlockMap {
'text': TextBlock
'reasoning': ReasoningBlock
'tool-call': ToolCallBlock
'tool-result': ToolResultBlock
'image': ImageBlock
}
export type ContentBlockType = keyof ContentBlockMap
export type ContentBlock = ContentBlockMap[ContentBlockType]
/** A single message in a conversation history. */
export interface Message {
role: 'system' | 'user' | 'assistant'
content: ContentBlock[]
}
/**
* Where a message (or injected content) came from.
* Merge-extensible sum type — plugins add their own `kind`s.
*/
export interface MessageSourceMap {
user: { kind: 'user' }
plugin: { kind: 'plugin'; plugin: string }
agent: { kind: 'agent'; agentId: string }
}
export type MessageSource = MessageSourceMap[keyof MessageSourceMap]
/**
* Why a model response stopped.
* Merge-extensible so adapters can surface provider-specific reasons.
*/
export interface FinishReasonMap {
'stop': { kind: 'stop' }
'tool-calls': { kind: 'tool-calls' }
'max-tokens': { kind: 'max-tokens' }
'aborted': { kind: 'aborted' }
'error': { kind: 'error'; message: string; code?: string }
}
export type FinishReason = FinishReasonMap[keyof FinishReasonMap]
/** Token accounting for one model call (cache fields are optional). */
export interface TokenUsage {
inputTokens: number
outputTokens: number
cacheReadTokens?: number
cacheWriteTokens?: number
reasoningTokens?: number
}
/**
* Raw streaming protocol emitted by adapters.
*
* A streaming response interleaves several typed blocks (text, reasoning,
* multiple tool calls); `index` ties each delta to its block, and `block-end`
* carries the fully-assembled ContentBlock so consumers don't have to
* re-assemble deltas themselves (use {@link BlockAssembler} when they do).
*
* TODO(review): this protocol needs careful review before the first real
* adapter lands (DeepSeek V4 wire format, partial JSON arguments, interleaved
* reasoning signatures, …).
*/
export type StreamChunk =
| { type: 'block-start'; index: number; blockType: ContentBlockType }
| { type: 'text-delta'; index: number; text: string }
| { type: 'reasoning-delta'; index: number; text: string }
| { type: 'tool-call-delta'; index: number; id: string; name?: string; argumentsDelta: string }
| { type: 'block-end'; index: number; block: ContentBlock }
| { type: 'usage'; usage: TokenUsage }
| { type: 'finish'; reason: FinishReason }
/**
* JSON-schema description of a tool, as sent to the model.
*
* Declared here (not in dsh-tools) because it is part of {@link GenerateOptions};
* dsh-tools' ToolDefinition and dsh-system-prompt's PromptAssembly both import
* it from this package.
*/
export interface ToolSchema {
name: string
description: string
/** JSON Schema object for the arguments. */
parameters: Record<string, unknown>
strict?: boolean
}
/** A single model request, fully assembled. */
export interface GenerateOptions {
model: string
messages: Message[]
/** System prompt text (adapters map to the provider's system slot). */
system?: string
/** Tool schemas (adapters map to the provider's `tools` field). */
tools?: ToolSchema[]
/** Assistant prefix continuation (prefill). */
prefill?: ContentBlock[]
temperature?: number
maxTokens?: number
stop?: string[]
signal?: AbortSignal
}
/** Non-streaming result, assembled from the chunk stream. */
export interface GenerateResult {
message: Message
usage?: TokenUsage
finish: FinishReason
}
+46
View File
@@ -0,0 +1,46 @@
import { describe, expect, it } from 'vitest'
import { BlockAssembler, type StreamChunk } from '@deepseek-ai/dsh-llm'
describe('BlockAssembler', () => {
it('assembles interleaved text, reasoning, and tool-call deltas', () => {
const chunks: StreamChunk[] = [
{ type: 'block-start', index: 0, blockType: 'reasoning' },
{ type: 'reasoning-delta', index: 0, text: 'thinking…' },
{ type: 'block-end', index: 0, block: { type: 'reasoning', text: 'thinking…' } },
{ type: 'block-start', index: 1, blockType: 'text' },
{ type: 'text-delta', index: 1, text: 'Hello' },
{ type: 'text-delta', index: 1, text: ' world' },
{ type: 'block-start', index: 2, blockType: 'tool-call' },
{ type: 'tool-call-delta', index: 2, id: 'call-1', name: 'echo', argumentsDelta: '{"text":' },
{ type: 'tool-call-delta', index: 2, id: 'call-1', argumentsDelta: '"hi"}' },
{ type: 'usage', usage: { inputTokens: 10, outputTokens: 5 } },
{ type: 'finish', reason: { kind: 'tool-calls' } },
]
const assembler = new BlockAssembler()
for (const chunk of chunks) assembler.push(chunk)
expect(assembler.blocks()).toEqual([
{ type: 'reasoning', text: 'thinking…' },
{ type: 'text', text: 'Hello world' },
{ type: 'tool-call', id: 'call-1', name: 'echo', arguments: '{"text":"hi"}' },
])
expect(assembler.usage).toEqual({ inputTokens: 10, outputTokens: 5 })
expect(assembler.finish).toEqual({ kind: 'tool-calls' })
expect(assembler.message().role).toBe('assistant')
})
it('returns the completed block from push() on block-end', () => {
const assembler = new BlockAssembler()
expect(assembler.push({ type: 'block-start', index: 0, blockType: 'text' })).toBeUndefined()
expect(assembler.push({ type: 'text-delta', index: 0, text: 'hi' })).toBeUndefined()
const block = assembler.push({ type: 'block-end', index: 0, block: { type: 'text', text: 'hi' } })
expect(block).toEqual({ type: 'text', text: 'hi' })
})
it('tolerates deltas without explicit block-start/end', () => {
const assembler = new BlockAssembler()
assembler.push({ type: 'text-delta', index: 0, text: 'implicit' })
expect(assembler.blocks()).toEqual([{ type: 'text', text: 'implicit' }])
expect(assembler.finish).toEqual({ kind: 'stop' })
})
})
+73
View File
@@ -0,0 +1,73 @@
import { describe, expect, it } from 'vitest'
import { Context } from 'cordis'
import LlmService, { GenerateOptions, LlmAdapter, StreamChunk } from '@deepseek-ai/dsh-llm'
class ScriptedAdapter extends LlmAdapter {
constructor(private script: StreamChunk[]) {
super()
}
async * stream(_options: GenerateOptions): AsyncIterable<StreamChunk> {
yield * this.script
}
}
const SCRIPT: StreamChunk[] = [
{ type: 'block-start', index: 0, blockType: 'text' },
{ type: 'text-delta', index: 0, text: 'hi' },
{ type: 'finish', reason: { kind: 'stop' } },
]
describe('LlmService', () => {
it('routes stream() to the registered adapter and generate() assembles it', async () => {
const ctx = new Context()
await ctx.plugin(LlmService)
ctx.llm.registerAdapter(['test-model'], new ScriptedAdapter(SCRIPT))
const chunks: StreamChunk[] = []
for await (const chunk of ctx.llm.stream({ model: 'test-model', messages: [] })) chunks.push(chunk)
expect(chunks).toHaveLength(3)
const result = await ctx.llm.generate({ model: 'test-model', messages: [] })
expect(result.message.content).toEqual([{ type: 'text', text: 'hi' }])
expect(result.finish).toEqual({ kind: 'stop' })
})
it('throws NO_ADAPTER for unregistered models', async () => {
const ctx = new Context()
await ctx.plugin(LlmService)
await expect(ctx.llm.generate({ model: 'nope', messages: [] })).rejects.toThrow('no adapter registered')
})
it('unregisters adapters when the owning fiber is disposed (HMR safety)', async () => {
const ctx = new Context()
await ctx.plugin(LlmService)
const fiber = await ctx.plugin(Object.assign((inner: Context) => {
inner.llm.registerAdapter(['scoped-model'], new ScriptedAdapter(SCRIPT))
}, { inject: ['llm'] }))
expect(ctx.llm.models()).toEqual(['scoped-model'])
await fiber.dispose()
expect(ctx.llm.models()).toEqual([])
})
it('lets llm/stream waterfall listeners wrap the underlying stream', async () => {
const ctx = new Context()
await ctx.plugin(LlmService)
ctx.llm.registerAdapter(['test-model'], new ScriptedAdapter(SCRIPT))
ctx.on('llm/stream', function (options, next) {
const inner = next()
return (async function * () {
yield { type: 'block-start', index: 99, blockType: 'text' } satisfies StreamChunk
yield * inner
})()
})
const chunks: StreamChunk[] = []
for await (const chunk of ctx.llm.stream({ model: 'test-model', messages: [] })) chunks.push(chunk)
expect(chunks).toHaveLength(4)
expect(chunks[0]).toMatchObject({ index: 99 })
})
})
+12
View File
@@ -0,0 +1,12 @@
{
"extends": "../../tsconfig.base.json",
"compilerOptions": {
"rootDir": "src",
"outDir": "lib"
},
"include": ["src"],
"references": [
{ "path": "../../vendor/cosmokit" },
{ "path": "../../vendor/cordis" }
]
}
+30
View File
@@ -0,0 +1,30 @@
{
"name": "@deepseek-ai/dsh-session",
"description": "Event-sourced session store for the DeepSeek Harness",
"version": "0.0.1",
"private": true,
"type": "module",
"main": "lib/index.js",
"types": "lib/index.d.ts",
"exports": {
".": {
"types": "./lib/index.d.ts",
"default": "./lib/index.js"
},
"./src/*": "./src/*",
"./package.json": "./package.json"
},
"files": [
"lib",
"src"
],
"license": "BSD-3-Clause",
"peerDependencies": {
"@deepseek-ai/dsh-llm": "^0.0.1",
"cordis": "^4.0.0-rc.6"
},
"devDependencies": {
"@deepseek-ai/dsh-llm": "^0.0.1",
"cordis": "^4.0.0-rc.6"
}
}
+163
View File
@@ -0,0 +1,163 @@
import { Context, Service } from 'cordis'
import type { ContentBlock, Message, MessageSource } from '@deepseek-ai/dsh-llm'
import type { SessionEvent, SessionEventMap, SessionEventType } from './types.ts'
export * from './types.ts'
declare module 'cordis' {
interface Context {
sessions: SessionStore
}
interface Events {
/** A session was created in the store. */
'session/created'(session: Session): void
/** An event was appended to a session log (sync, fire-and-forget). */
'session/event'(session: Session, event: SessionEvent): void
/**
* Awaited durability checkpoint. The agent loop awaits
* `ctx.parallel('session/flush', session)` at every turn end; persistence
* plugins (JSONL, sqlite — TODO, future phase) drain their write-behind
* buffers here and on fiber dispose.
*/
'session/flush'(session: Session): Promise<void> | void
}
}
/**
* Renders a `context/message` or `steering/message` event as a tagged
* synthetic user-role message (the system-reminder pattern: zero adapter
* burden, models distinguish it from real user prompts by the envelope).
*
* TODO(review): revisit the envelope once a real adapter exists.
*/
function renderTagged(tag: string, content: ContentBlock[], source: MessageSource): ContentBlock[] {
const open = `<${tag} source=${JSON.stringify(source.kind)}>`
const close = `</${tag}>`
return [
{ type: 'text', text: open },
...content,
{ type: 'text', text: close },
]
}
/**
* An event-sourced session: an append-only log of {@link SessionEvent}s.
*
* Plain class (not a Service) — create instances via `ctx.sessions.create()`.
* Seeding with an existing event log replays/forks a session.
*/
export class Session {
private log: SessionEvent[] = []
/** Set by the store so appends are observable; no-op when detached. */
onAppend?: (event: SessionEvent) => void
constructor(public readonly id: string, seed?: SessionEvent[]) {
if (seed) this.log = [...seed]
}
get events(): readonly SessionEvent[] {
return this.log
}
get seq(): number {
return this.log.length
}
/** Append one event. Synchronous — the hot path never blocks on I/O. */
append<T extends SessionEventType>(type: T, data: SessionEventMap[T]): SessionEvent<T> {
const event: SessionEvent<T> = { type, seq: this.log.length, time: Date.now(), data }
this.log.push(event as SessionEvent)
this.onAppend?.(event as SessionEvent)
return event
}
/**
* Derive the LLM message history from the event log.
*
* - `user/message` → user message
* - `assistant/message` → assistant message (chunks are skipped — they are
* replay/UI data; the assembled message is authoritative for history)
* - `tool/result` → user message carrying a tool-result block
* - `context/message` / `steering/message` → tagged synthetic user messages
* at their chronological position
*/
deriveMessages(): Message[] {
const messages: Message[] = []
for (const event of this.log) {
switch (event.type) {
case 'user/message': {
const { content } = event.data as SessionEventMap['user/message']
messages.push({ role: 'user', content })
break
}
case 'assistant/message': {
const { content } = event.data as SessionEventMap['assistant/message']
messages.push({ role: 'assistant', content })
break
}
case 'tool/result': {
const { callId, content, isError } = event.data as SessionEventMap['tool/result']
messages.push({
role: 'user',
content: [{ type: 'tool-result', toolCallId: callId, content, isError }],
})
break
}
case 'context/message': {
const { content, source } = event.data as SessionEventMap['context/message']
messages.push({ role: 'user', content: renderTagged('context', content, source) })
break
}
case 'steering/message': {
const { content, source } = event.data as SessionEventMap['steering/message']
messages.push({ role: 'user', content: renderTagged('steering', content, source) })
break
}
}
}
return messages
}
}
/**
* In-memory session store (`ctx.sessions`).
*
* Persistence is intentionally not implemented here — persistence plugins
* subscribe to `session/event` and flush on `session/flush` / dispose.
*/
export class SessionStore extends Service {
private store = new Map<string, Session>()
private counter = 0
constructor(ctx: Context) {
super(ctx, 'sessions')
}
/** Create a session. `seed` replays/forks an existing event log. */
create(id?: string, seed?: SessionEvent[]): Session {
id ??= `session-${++this.counter}`
if (this.store.has(id)) throw new Error(`session "${id}" already exists`)
const session = new Session(id, seed)
session.onAppend = (event) => this.ctx.emit('session/event', session, event)
this.ctx.effect(() => {
this.store.set(id, session)
this.ctx.emit('session/created', session)
return () => {
session.onAppend = undefined
this.store.delete(id)
}
}, 'sessions.create()')
return session
}
get(id: string): Session | undefined {
return this.store.get(id)
}
list(): Session[] {
return [...this.store.values()]
}
}
export default SessionStore
+74
View File
@@ -0,0 +1,74 @@
import type { ContentBlock, MessageSource, StreamChunk, TokenUsage } from '@deepseek-ai/dsh-llm'
/**
* What started a turn.
* Merge-extensible sum type (same pattern as MessageSourceMap).
*/
export interface TurnTriggerMap {
message: { kind: 'message'; source: MessageSource }
continuation: { kind: 'continuation' }
}
export type TurnTrigger = TurnTriggerMap[keyof TurnTriggerMap]
/**
* Why a turn ended.
* Merge-extensible sum type.
*/
export interface TurnEndReasonMap {
completed: { kind: 'completed' }
aborted: { kind: 'aborted'; reason?: string }
error: { kind: 'error'; message: string; code?: string }
disposed: { kind: 'disposed' }
}
export type TurnEndReason = TurnEndReasonMap[keyof TurnEndReasonMap]
/**
* The session event vocabulary — the append-only source of truth for an
* agent's whole interaction history. The LLM message history is *derived*
* from this log; nothing else is authoritative. Replay = re-derive from the
* same events; trace/telemetry = subscribe to the log.
*
* Merge-extensible: plugins declare extra event types via declaration merging
* (e.g. a compaction plugin adds `'compaction/marker'`).
*
* TODO(review): this vocabulary needs careful review once the loop and the
* first persistence plugin exist side by side.
*/
export interface SessionEventMap {
'turn/start': { turn: number; trigger: TurnTrigger }
'turn/end': { turn: number; reason: TurnEndReason }
'step/start': { turn: number; step: number }
'step/end': { turn: number; step: number }
/** A user-visible prompt (queued message drained at turn start). */
'user/message': { content: ContentBlock[]; source: MessageSource }
/**
* In-session context injection (file-change notices, subdir AGENTS.md,
* skill content, cron notifications, …). Rendered into the derived history
* as tagged synthetic context — NOT a user prompt.
*/
'context/message': { content: ContentBlock[]; source: MessageSource }
/** Raw stream chunk — token-level replay fidelity. */
'assistant/chunk': { turn: number; step: number; chunk: StreamChunk }
/** Assembled assistant message for one step (derived history uses this). */
'assistant/message': { turn: number; step: number; content: ContentBlock[] }
'tool/call': { turn: number; step: number; callId: string; name: string; arguments: string }
'tool/result': { turn: number; step: number; callId: string; content: ContentBlock[]; isError: boolean }
/** Steering content injected between steps of a running turn. */
'steering/message': { turn: number; content: ContentBlock[]; source: MessageSource }
'usage': { turn: number; step: number; usage: TokenUsage }
'error': { turn: number; step: number; message: string; code?: string }
}
export type SessionEventType = keyof SessionEventMap
/** One immutable entry in the session log. */
export interface SessionEvent<T extends SessionEventType = SessionEventType> {
type: T
/** Monotonic sequence number within the session. */
seq: number
/** Unix epoch milliseconds. */
time: number
data: SessionEventMap[T]
}
+109
View File
@@ -0,0 +1,109 @@
import { describe, expect, it } from 'vitest'
import { Context } from 'cordis'
import SessionStore, { Session, SessionEvent } from '@deepseek-ai/dsh-session'
describe('Session', () => {
it('derives message history from the event log', () => {
const session = new Session('s1')
session.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
session.append('user/message', { content: [{ type: 'text', text: 'hello' }], source: { kind: 'user' } })
session.append('assistant/chunk', { turn: 1, step: 1, chunk: { type: 'text-delta', index: 0, text: 'hi' } })
session.append('assistant/message', {
turn: 1, step: 1,
content: [
{ type: 'text', text: 'let me check' },
{ type: 'tool-call', id: 'c1', name: 'echo', arguments: '{}' },
],
})
session.append('tool/result', { turn: 1, step: 1, callId: 'c1', content: [{ type: 'text', text: 'ok' }], isError: false })
session.append('turn/end', { turn: 1, reason: { kind: 'completed' } })
const messages = session.deriveMessages()
expect(messages.map(m => m.role)).toEqual(['user', 'assistant', 'user'])
// raw chunks must NOT appear in derived history
expect(messages[1].content).toHaveLength(2)
expect(messages[2].content[0]).toMatchObject({ type: 'tool-result', toolCallId: 'c1' })
})
it('renders context and steering messages as tagged synthetic user content', () => {
const session = new Session('s2')
session.append('context/message', {
content: [{ type: 'text', text: 'file changed: a.ts' }],
source: { kind: 'plugin', plugin: 'watcher' },
})
session.append('steering/message', {
turn: 1,
content: [{ type: 'text', text: 'focus on tests' }],
source: { kind: 'user' },
})
const [contextMessage, steeringMessage] = session.deriveMessages()
expect(contextMessage.role).toBe('user')
expect(contextMessage.content[0]).toMatchObject({ type: 'text', text: '<context source="plugin">' })
expect(contextMessage.content.at(-1)).toMatchObject({ type: 'text', text: '</context>' })
expect(steeringMessage.content[0]).toMatchObject({ type: 'text', text: '<steering source="user">' })
})
it('replays identically from a seeded event log', () => {
const original = new Session('s3')
original.append('user/message', { content: [{ type: 'text', text: 'q' }], source: { kind: 'user' } })
original.append('assistant/message', { turn: 1, step: 1, content: [{ type: 'text', text: 'a' }] })
const replayed = new Session('s3-replay', [...original.events])
expect(replayed.deriveMessages()).toEqual(original.deriveMessages())
expect(replayed.seq).toBe(original.seq)
})
})
describe('SessionStore', () => {
it('creates sessions, emits session/created and session/event', async () => {
const ctx = new Context()
await ctx.plugin(SessionStore)
const created: Session[] = []
const events: [Session, SessionEvent][] = []
ctx.on('session/created', session => void created.push(session))
ctx.on('session/event', (session, event) => void events.push([session, event]))
const session = ctx.sessions.create()
expect(created).toEqual([session])
session.append('user/message', { content: [{ type: 'text', text: 'x' }], source: { kind: 'user' } })
expect(events).toHaveLength(1)
expect(events[0][0]).toBe(session)
expect(events[0][1].type).toBe('user/message')
expect(ctx.sessions.get(session.id)).toBe(session)
expect(ctx.sessions.list()).toEqual([session])
})
it('rejects duplicate ids and supports seeding', async () => {
const ctx = new Context()
await ctx.plugin(SessionStore)
const a = ctx.sessions.create('fixed')
expect(() => ctx.sessions.create('fixed')).toThrow('already exists')
a.append('user/message', { content: [{ type: 'text', text: 'q' }], source: { kind: 'user' } })
const forked = ctx.sessions.create('fork', [...a.events])
expect(forked.deriveMessages()).toEqual(a.deriveMessages())
})
it('detaches sessions when the creating fiber is disposed (HMR safety)', async () => {
const ctx = new Context()
await ctx.plugin(SessionStore)
let session!: Session
const fiber = await ctx.plugin(Object.assign((inner: Context) => {
session = inner.sessions.create('scoped')
}, { inject: ['sessions'] }))
expect(ctx.sessions.get('scoped')).toBe(session)
let observed = 0
ctx.on('session/event', () => void observed++)
await fiber.dispose()
expect(ctx.sessions.get('scoped')).toBeUndefined()
session.append('user/message', { content: [{ type: 'text', text: 'late' }], source: { kind: 'user' } })
expect(observed).toBe(0)
})
})
+13
View File
@@ -0,0 +1,13 @@
{
"extends": "../../tsconfig.base.json",
"compilerOptions": {
"rootDir": "src",
"outDir": "lib"
},
"include": ["src"],
"references": [
{ "path": "../../vendor/cosmokit" },
{ "path": "../../vendor/cordis" },
{ "path": "../llm" }
]
}
+30
View File
@@ -0,0 +1,30 @@
{
"name": "@deepseek-ai/dsh-system-prompt",
"description": "System prompt assembly registry for the DeepSeek Harness",
"version": "0.0.1",
"private": true,
"type": "module",
"main": "lib/index.js",
"types": "lib/index.d.ts",
"exports": {
".": {
"types": "./lib/index.d.ts",
"default": "./lib/index.js"
},
"./src/*": "./src/*",
"./package.json": "./package.json"
},
"files": [
"lib",
"src"
],
"license": "BSD-3-Clause",
"peerDependencies": {
"@deepseek-ai/dsh-llm": "^0.0.1",
"cordis": "^4.0.0-rc.6"
},
"devDependencies": {
"@deepseek-ai/dsh-llm": "^0.0.1",
"cordis": "^4.0.0-rc.6"
}
}
+98
View File
@@ -0,0 +1,98 @@
import { Context, Service } from 'cordis'
import type { ToolSchema } from '@deepseek-ai/dsh-llm'
declare module 'cordis' {
interface Context {
systemPrompt: SystemPrompt
}
interface Events {
/** Waterfall around prompt assembly — mutate/extend the assembly. */
'system-prompt/assemble'(this: SystemPrompt, assembly: PromptAssembly, next: () => Promise<PromptAssembly>): Promise<PromptAssembly>
/** A section or tool provider was registered or unregistered. */
'system-prompt/change'(): void
}
}
/** One contributed section of the system prompt. */
export interface PromptSection {
/** Unique name (diagnostics / dedup). */
name: string
/** Sections are concatenated in ascending order. */
order: number
/** Static text or a provider evaluated at each assembly. */
text: string | (() => string)
}
/**
* The assembled prompt.
*
* Tool schemas are part of the assembly by design: "what the model is told it
* can do" is one coherent thing managed here, even though adapters transmit
* `tools` as a separate wire field rather than prompt text.
*
* Merge-extensible: plugins can declare extra fields on this interface.
*/
export interface PromptAssembly {
sections: PromptSection[]
tools: ToolSchema[]
}
/** Renders the text part of an assembly (sections joined by blank lines). */
export function renderPrompt(assembly: PromptAssembly): string {
return assembly.sections
.map(section => typeof section.text === 'function' ? section.text() : section.text)
.filter(text => text.length > 0)
.join('\n\n')
}
/**
* Registry service (`ctx.systemPrompt`): plugins contribute ordered text
* sections and tool-schema providers; the agent loop calls `assemble()` once
* per step.
*/
export class SystemPrompt extends Service {
private sections: PromptSection[] = []
private toolProviders: (() => ToolSchema[])[] = []
constructor(ctx: Context) {
super(ctx, 'systemPrompt')
}
/** Contribute a section. Disposed with the calling fiber. */
section(section: PromptSection): () => void {
return this.ctx.effect(() => {
this.sections.push(section)
this.ctx.emit('system-prompt/change')
return () => {
const index = this.sections.indexOf(section)
if (index >= 0) this.sections.splice(index, 1)
this.ctx.emit('system-prompt/change')
}
}, 'systemPrompt.section()')
}
/** Contribute tool schemas (evaluated at each assembly). Disposed with the fiber. */
tools(provider: () => ToolSchema[]): () => void {
return this.ctx.effect(() => {
this.toolProviders.push(provider)
this.ctx.emit('system-prompt/change')
return () => {
const index = this.toolProviders.indexOf(provider)
if (index >= 0) this.toolProviders.splice(index, 1)
this.ctx.emit('system-prompt/change')
}
}, 'systemPrompt.tools()')
}
/** Assemble the current prompt (sections sorted, tools collected). */
assemble(): Promise<PromptAssembly> {
const assembly: PromptAssembly = {
sections: [...this.sections].sort((a, b) => a.order - b.order),
tools: this.toolProviders.flatMap(provider => provider()),
}
return this.ctx.waterfall(this, 'system-prompt/assemble', assembly, async () => assembly)
}
}
export default SystemPrompt
@@ -0,0 +1,71 @@
import { describe, expect, it } from 'vitest'
import { Context } from 'cordis'
import SystemPrompt, { PromptAssembly, renderPrompt } from '@deepseek-ai/dsh-system-prompt'
describe('SystemPrompt', () => {
it('assembles sections in order with dynamic text and collected tools', async () => {
const ctx = new Context()
await ctx.plugin(SystemPrompt)
ctx.systemPrompt.section({ name: 'persona', order: 0, text: 'You are DeepSeek Code.' })
ctx.systemPrompt.section({ name: 'cwd', order: 20, text: () => `cwd: /tmp` })
ctx.systemPrompt.section({ name: 'rules', order: 10, text: 'Be precise.' })
ctx.systemPrompt.tools(() => [{ name: 'echo', description: 'echo back', parameters: {} }])
const assembly = await ctx.systemPrompt.assemble()
expect(assembly.sections.map(s => s.name)).toEqual(['persona', 'rules', 'cwd'])
expect(assembly.tools).toEqual([{ name: 'echo', description: 'echo back', parameters: {} }])
expect(renderPrompt(assembly)).toBe('You are DeepSeek Code.\n\nBe precise.\n\ncwd: /tmp')
})
it('removes contributions when the contributing fiber is disposed (HMR safety)', async () => {
const ctx = new Context()
await ctx.plugin(SystemPrompt)
const fiber = await ctx.plugin(Object.assign((inner: Context) => {
inner.systemPrompt.section({ name: 'scoped', order: 0, text: 'scoped section' })
inner.systemPrompt.tools(() => [{ name: 'scoped-tool', description: '', parameters: {} }])
}, { inject: ['systemPrompt'] }))
expect((await ctx.systemPrompt.assemble()).sections).toHaveLength(1)
await fiber.dispose()
const assembly = await ctx.systemPrompt.assemble()
expect(assembly.sections).toHaveLength(0)
expect(assembly.tools).toHaveLength(0)
})
it('composes multiple system-prompt/assemble waterfall listeners in order', async () => {
const ctx = new Context()
await ctx.plugin(SystemPrompt)
ctx.systemPrompt.section({ name: 'base', order: 0, text: 'base' })
// Listener A appends a section, then delegates.
ctx.on('system-prompt/assemble', async (assembly: PromptAssembly, next) => {
assembly.sections.push({ name: 'from-a', order: 100, text: 'a' })
return next()
})
// Listener B (registered later, runs after A) sees A's contribution.
const seen: string[][] = []
ctx.on('system-prompt/assemble', async (assembly: PromptAssembly, next) => {
seen.push(assembly.sections.map(s => s.name))
return next()
})
const assembly = await ctx.systemPrompt.assemble()
expect(seen).toEqual([['base', 'from-a']])
expect(assembly.sections.map(s => s.name)).toEqual(['base', 'from-a'])
})
it('lets a waterfall listener short-circuit by not calling next()', async () => {
const ctx = new Context()
await ctx.plugin(SystemPrompt)
ctx.systemPrompt.section({ name: 'real', order: 0, text: 'real' })
ctx.on('system-prompt/assemble', async () => {
return { sections: [], tools: [] } satisfies PromptAssembly
})
const assembly = await ctx.systemPrompt.assemble()
expect(assembly.sections).toHaveLength(0)
})
})
+13
View File
@@ -0,0 +1,13 @@
{
"extends": "../../tsconfig.base.json",
"compilerOptions": {
"rootDir": "src",
"outDir": "lib"
},
"include": ["src"],
"references": [
{ "path": "../../vendor/cosmokit" },
{ "path": "../../vendor/cordis" },
{ "path": "../llm" }
]
}
+34
View File
@@ -0,0 +1,34 @@
{
"name": "@deepseek-ai/dsh-tools",
"description": "Tool registry and execution pipeline for the DeepSeek Harness",
"version": "0.0.1",
"private": true,
"type": "module",
"main": "lib/index.js",
"types": "lib/index.d.ts",
"exports": {
".": {
"types": "./lib/index.d.ts",
"default": "./lib/index.js"
},
"./src/*": "./src/*",
"./package.json": "./package.json"
},
"files": [
"lib",
"src"
],
"license": "BSD-3-Clause",
"peerDependencies": {
"@deepseek-ai/dsh-agent": "^0.0.1",
"@deepseek-ai/dsh-llm": "^0.0.1",
"@deepseek-ai/dsh-system-prompt": "^0.0.1",
"cordis": "^4.0.0-rc.6"
},
"devDependencies": {
"@deepseek-ai/dsh-agent": "^0.0.1",
"@deepseek-ai/dsh-llm": "^0.0.1",
"@deepseek-ai/dsh-system-prompt": "^0.0.1",
"cordis": "^4.0.0-rc.6"
}
}
+117
View File
@@ -0,0 +1,117 @@
import { Context, Service } from 'cordis'
import type { ContentBlock, ToolSchema } from '@deepseek-ai/dsh-llm'
import type { Agent } from '@deepseek-ai/dsh-agent'
import type {} from '@deepseek-ai/dsh-system-prompt'
declare module 'cordis' {
interface Context {
tools: ToolRegistry
}
interface Events {
/**
* Waterfall around every tool execution — the single seam where sandbox,
* permission, hook, and plan-mode plugins wrap or veto a call. Listeners
* receive `(exec, next)`: call `next()` to proceed (possibly around your
* own logic), or return a ToolExecutionResult without calling `next()`
* to short-circuit (veto).
*/
'tools/execute'(this: ToolRegistry, exec: ToolExecution, next: () => Promise<ToolExecutionResult>): Promise<ToolExecutionResult>
/** A tool was registered or unregistered. */
'tools/change'(): void
}
}
// TODO(review): revisit these shapes when the first real tools and
// sandbox/permission plugins land (e.g. a concurrency-safety hint for
// parallel execution — Claude Code partitions read-only tools; phase 1
// executes sequentially).
/** A registered tool: its schema plus the execution function. */
export interface ToolDefinition extends ToolSchema {
execute(args: unknown, exec: ToolExecution): Promise<ContentBlock[]>
}
/** One pending tool call, as it flows through the execution waterfall. */
export interface ToolExecution {
callId: string
name: string
/** Parsed JSON arguments (unknown — tools validate their own input). */
arguments: unknown
/** The agent on whose behalf the call runs (set by the agent loop). */
agent?: Agent
signal?: AbortSignal
}
/** The outcome of one tool call. */
export interface ToolExecutionResult {
callId: string
content: ContentBlock[]
isError: boolean
}
/**
* Tool registry (`ctx.tools`): tool plugins register definitions; the agent
* loop executes calls through the `tools/execute` waterfall. The registry
* contributes its schemas into the system-prompt assembly.
*/
export class ToolRegistry extends Service {
static inject = ['systemPrompt']
private store = new Map<string, ToolDefinition>()
constructor(ctx: Context) {
super(ctx, 'tools')
ctx.systemPrompt.tools(() => this.schemas())
}
/** Register a tool. Disposed with the calling fiber. */
register(definition: ToolDefinition): () => void {
return this.ctx.effect(() => {
if (this.store.has(definition.name)) {
throw new Error(`tool "${definition.name}" is already registered`)
}
this.store.set(definition.name, definition)
this.ctx.emit('tools/change')
return () => {
this.store.delete(definition.name)
this.ctx.emit('tools/change')
}
}, 'tools.register()')
}
get(name: string): ToolDefinition | undefined {
return this.store.get(name)
}
/** Schemas of all registered tools (without the execute functions). */
schemas(): ToolSchema[] {
return [...this.store.values()].map(({ execute, ...schema }) => schema)
}
/** Execute one tool call through the `tools/execute` waterfall. */
execute(exec: ToolExecution): Promise<ToolExecutionResult> {
return this.ctx.waterfall(this, 'tools/execute', exec, async (): Promise<ToolExecutionResult> => {
const tool = this.store.get(exec.name)
if (!tool) {
return {
callId: exec.callId,
content: [{ type: 'text', text: `Error: unknown tool "${exec.name}"` }],
isError: true,
}
}
try {
const content = await tool.execute(exec.arguments, exec)
return { callId: exec.callId, content, isError: false }
} catch (error: any) {
return {
callId: exec.callId,
content: [{ type: 'text', text: `Error: ${error?.message ?? error}` }],
isError: true,
}
}
})
}
}
export default ToolRegistry
+120
View File
@@ -0,0 +1,120 @@
import { describe, expect, it } from 'vitest'
import { Context } from 'cordis'
import SystemPrompt from '@deepseek-ai/dsh-system-prompt'
import ToolRegistry, { ToolExecutionResult } from '@deepseek-ai/dsh-tools'
async function setup() {
const ctx = new Context()
await ctx.plugin(SystemPrompt)
await ctx.plugin(ToolRegistry)
return ctx
}
const echoTool = {
name: 'echo',
description: 'echo arguments back',
parameters: { type: 'object', properties: { text: { type: 'string' } } },
async execute(args: any) {
return [{ type: 'text' as const, text: String(args?.text ?? '') }]
},
}
describe('ToolRegistry', () => {
it('registers tools, exposes schemas, and feeds the system-prompt assembly', async () => {
const ctx = await setup()
ctx.tools.register(echoTool)
expect(ctx.tools.schemas()).toEqual([{
name: 'echo',
description: 'echo arguments back',
parameters: { type: 'object', properties: { text: { type: 'string' } } },
}])
// schemas() result must not leak execute
expect((ctx.tools.schemas()[0] as any).execute).toBeUndefined()
const assembly = await ctx.systemPrompt.assemble()
expect(assembly.tools.map(t => t.name)).toEqual(['echo'])
})
it('executes a tool and returns its content', async () => {
const ctx = await setup()
ctx.tools.register(echoTool)
const result = await ctx.tools.execute({ callId: 'c1', name: 'echo', arguments: { text: 'hi' } })
expect(result).toEqual({ callId: 'c1', content: [{ type: 'text', text: 'hi' }], isError: false })
})
it('returns isError results for unknown tools and throwing tools', async () => {
const ctx = await setup()
ctx.tools.register({
...echoTool,
name: 'boom',
async execute() {
throw new Error('exploded')
},
})
const unknown = await ctx.tools.execute({ callId: 'c1', name: 'nope', arguments: {} })
expect(unknown.isError).toBe(true)
const thrown = await ctx.tools.execute({ callId: 'c2', name: 'boom', arguments: {} })
expect(thrown.isError).toBe(true)
expect(thrown.content[0]).toMatchObject({ text: 'Error: exploded' })
})
it('lets tools/execute waterfall listeners veto a call (permission pattern)', async () => {
const ctx = await setup()
ctx.tools.register(echoTool)
ctx.on('tools/execute', async (exec, next): Promise<ToolExecutionResult> => {
if (exec.name === 'echo') {
return {
callId: exec.callId,
content: [{ type: 'text', text: 'denied by policy' }],
isError: true,
}
}
return next()
})
const result = await ctx.tools.execute({ callId: 'c1', name: 'echo', arguments: { text: 'hi' } })
expect(result.isError).toBe(true)
expect(result.content[0]).toMatchObject({ text: 'denied by policy' })
})
it('composes multiple tools/execute listeners (sandbox-wrap pattern)', async () => {
const ctx = await setup()
ctx.tools.register(echoTool)
const order: string[] = []
ctx.on('tools/execute', async (exec, next) => {
order.push('first:before')
const result = await next()
order.push('first:after')
return result
})
ctx.on('tools/execute', async (exec, next) => {
order.push('second:before')
const result = await next()
order.push('second:after')
return result
})
const result = await ctx.tools.execute({ callId: 'c1', name: 'echo', arguments: { text: 'x' } })
expect(result.isError).toBe(false)
expect(order).toEqual(['first:before', 'second:before', 'second:after', 'first:after'])
})
it('rejects duplicate names and unregisters on fiber dispose (HMR safety)', async () => {
const ctx = await setup()
ctx.tools.register(echoTool)
expect(() => ctx.tools.register(echoTool)).toThrow('already registered')
const fiber = await ctx.plugin(Object.assign((inner: Context) => {
inner.tools.register({ ...echoTool, name: 'scoped' })
}, { inject: ['tools'] }))
expect(ctx.tools.schemas().map(t => t.name)).toEqual(['echo', 'scoped'])
await fiber.dispose()
expect(ctx.tools.schemas().map(t => t.name)).toEqual(['echo'])
})
})
+15
View File
@@ -0,0 +1,15 @@
{
"extends": "../../tsconfig.base.json",
"compilerOptions": {
"rootDir": "src",
"outDir": "lib"
},
"include": ["src"],
"references": [
{ "path": "../../vendor/cosmokit" },
{ "path": "../../vendor/cordis" },
{ "path": "../llm" },
{ "path": "../system-prompt" },
{ "path": "../agent" }
]
}