feat(acp): advertise and switch llm models

This commit is contained in:
Yichen Jiang
2026-07-15 13:33:42 +08:00
parent 75331f03a6
commit f1d39921c9
94 changed files with 1657 additions and 191 deletions

View File

@@ -1,6 +1,6 @@
# @deepseek-ai/dsh-llm-replay
A replay LLM plugin for keyless snapshot tests. It installs a single `llm/stream` waterfall listener that short-circuits the waterfall (never calls `next()`) and yields model streams reconstructed from a recorded **session JSONL** fixture — so a test can boot the real agent against a fixed model transcript with no API key.
A replay LLM plugin for keyless snapshot tests. It yields model streams reconstructed from a recorded **session JSONL** fixture, so a test can boot the real agent against a fixed model transcript with no API key. With `providers` configured it registers a replay-only adapter whose catalog is visible to clients such as ACP editors; without `providers` it installs the catch-all `llm/stream` waterfall used by tests that do not need discovery.
Its consumer is the ACP snapshot harness in `examples/acp-agent`, which loads this plugin (via `cordis.snapshot.yml`) in place of a real LLM adapter. The package exists so its derive/parse/replay logic falls under the per-file 100% coverage gate on `packages/*/src` (the same logic, while it lived under `examples/`, was outside the gate).
@@ -23,10 +23,18 @@ Replay keys every call by its calling session id (`GenerateOptions.sessionId`, s
| `file` | string | `$DSH_SNAPSHOT_FILE` | Path to the primary (parent) `session.jsonl` fixture. Required (config or env). |
| `overrideFile` | string | `$DSH_SNAPSHOT_OVERRIDE` | Optional path to a `ReplayEntry[]` sidecar that replaces the PRIMARY session's derived script. |
| `childFiles` | string[] | `$DSH_SNAPSHOT_CHILD_FILES` (path-delimited) | Recorded subagent child-session logs for a nested scenario; empty for a single-session scenario. |
| `providers` | `ReplayProviderConfig[]` | — | Optional replay-only provider and model catalog. Configured routes dispatch through the replay adapter and never perform provider I/O. |
```yaml
- id: llm-replay
name: '@deepseek-ai/dsh-llm-replay'
config:
providers:
- id: deepseek
name: DeepSeek
models:
- id: deepseek-v4-flash
- id: deepseek-v4-pro
# file/overrideFile/childFiles default to $DSH_SNAPSHOT_FILE /
# $DSH_SNAPSHOT_OVERRIDE / $DSH_SNAPSHOT_CHILD_FILES, set by the snapshot
# harness per scenario.
@@ -34,11 +42,11 @@ Replay keys every call by its calling session id (`GenerateOptions.sessionId`, s
## Exports
- `installLlmReplay(ctx, config)` — install the `llm/stream` listener; returns the disposer (HMR safety). Use this in tests to drive replay without the Loader or env vars.
- `installLlmReplay(ctx, config)` — install the configured replay adapter or catch-all `llm/stream` listener; returns the disposer (HMR safety). Use this in tests to drive replay without the Loader or env vars.
- `loadSessionScripts(config)` — resolve the ordered `SessionScript[]` (primary + children) for a scenario, ready to bind to live sessions in first-call order.
- `loadReplayScript(config)` — resolve the `ReplayEntry[]` for the PRIMARY session only (sidecar override if present, else derived from the JSONL; fail-loud if the fixture is missing).
- `deriveReplayScript(events)` / `parseSessionLog(text)` / `parseSessionHeader(text)` — the pure helpers that turn a recorded session log into a script and read its header `id`/`createdAt`. A derived group must end in a `finish` chunk; a group without one is the fingerprint of a thrown `stream()` and must instead be expressed via an override sidecar.
- Types `ReplayEntry` / `SessionScript` / `ReplayConfig` / `Config`.
- Types `ReplayEntry` / `SessionScript` / `ReplayConfig` / `ReplayProviderConfig` / `ReplayModelConfig` / `Config`.
## Plugin export shape

View File

@@ -10,8 +10,8 @@ import { existsSync, readFileSync } from 'node:fs'
import { delimiter as pathDelimiter } from 'node:path'
import type { Context } from 'cordis'
import type { SessionEvent } from '@deepseek-ai/dsh-session'
import type { GenerateOptions, StreamChunk } from '@deepseek-ai/dsh-llm'
import { LlmError, assertNever } from '@deepseek-ai/dsh-llm'
import type { GenerateOptions, LlmModelInfo, LlmProviderInfo, StreamChunk } from '@deepseek-ai/dsh-llm'
import { LlmAdapter, LlmError, assertNever } from '@deepseek-ai/dsh-llm'
/**
* One recorded model call. `throw` may replay prefix chunks before failing;
@@ -23,6 +23,26 @@ export type ReplayEntry =
| { kind: 'throw'; chunks: StreamChunk[]; message: string; code: string; status?: number }
| { kind: 'hang' }
/** One model exposed by a replay-only provider catalog. */
export interface ReplayModelConfig {
/** Model id used for replay requests. */
id: string
/** Selector label; defaults to {@link id}. */
name?: string
/** Optional selector description. */
description?: string
}
/** One provider route exposed by the replay adapter. */
export interface ReplayProviderConfig {
/** Provider route used for replay requests. */
id: string
/** Selector label; defaults to {@link id}. */
name?: string
/** Advisory models exposed to clients such as ACP editors. */
models?: ReplayModelConfig[]
}
/** Resolved plugin configuration. */
export interface ReplayConfig {
/**
@@ -45,6 +65,12 @@ export interface ReplayConfig {
* for a single-session scenario.
*/
childFiles?: string[]
/**
* Optional provider catalog. When non-empty, replay registers an adapter for
* these routes; when absent or empty, it retains the catch-all waterfall used
* by tests that do not need discovery.
*/
providers?: ReplayProviderConfig[]
}
/**
@@ -203,6 +229,42 @@ export function loadSessionScripts(config: ReplayConfig): SessionScript[] {
return [primary, ...children]
}
/** Replay adapter that makes a configured provider catalog discoverable without provider I/O. */
class ReplayAdapter extends LlmAdapter {
private readonly providers: ReadonlyMap<string, ReplayProviderConfig>
constructor(
providers: readonly ReplayProviderConfig[],
private readonly replay: (options: GenerateOptions) => AsyncIterable<StreamChunk>,
) {
super()
this.providers = new Map(providers.map(provider => [provider.id, provider]))
}
override providerInfo(provider: string): LlmProviderInfo {
const configured = this.providers.get(provider)
/* v8 ignore next -- LlmService only asks about routes registered from this same map. */
if (configured === undefined) return super.providerInfo(provider)
return { id: provider, name: configured.name ?? provider }
}
override listModels(provider: string): Promise<readonly LlmModelInfo[]> {
const configured = this.providers.get(provider)
/* v8 ignore next -- LlmService only asks about routes registered from this same map. */
if (configured === undefined) return Promise.resolve([])
return Promise.resolve((configured.models ?? []).map(model => ({
provider,
id: model.id,
name: model.name ?? model.id,
...model.description === undefined ? {} : { description: model.description },
})))
}
override stream(options: GenerateOptions): AsyncIterable<StreamChunk> {
return this.replay(options)
}
}
/** Yield a recorded stream back, honoring abort like a real adapter. */
async function* replayEntry(entry: ReplayEntry, signal: AbortSignal | undefined): AsyncIterable<StreamChunk> {
switch (entry.kind) {
@@ -243,12 +305,14 @@ async function* replayEntry(entry: ReplayEntry, signal: AbortSignal | undefined)
/**
* Install per-session positional replay. A newly seen live session takes the
* next ordered recorded script, then advances its own cursor synchronously at
* invocation time; calls without `sessionId` share one anonymous session.
* Returns the effect disposer for HMR-safe removal.
* invocation time; calls without `sessionId` share one anonymous session. A
* non-empty provider catalog registers a routed replay adapter; otherwise a
* catch-all waterfall intercepts requests. Returns the effect disposer for
* HMR-safe removal.
*
* @param ctx - the context whose `llm/stream` waterfall the listener short-circuits.
* @param ctx - the context whose LLM service receives the replay route or waterfall.
* @param config - the resolved fixture paths (env-var defaulting is `apply`'s job).
* @returns the `ctx.on` disposer that removes the listener.
* @returns the disposer that removes the registered adapter or listener.
*/
export function installLlmReplay(ctx: Context, config: ReplayConfig): () => void {
const scripts = loadSessionScripts(config)
@@ -258,7 +322,7 @@ export function installLlmReplay(ctx: Context, config: ReplayConfig): () => void
const bound = new Map<string, { entries: ReplayEntry[]; cursor: number }>()
let nextScript = 0
const ANON = '\0anon\0' // the key for a call that carries no sessionId
return ctx.on('llm/stream', (options: GenerateOptions, _next) => {
const replay = (options: GenerateOptions): AsyncIterable<StreamChunk> => {
const key = options.sessionId ?? ANON
let state = bound.get(key)
let unrecorded = false
@@ -296,7 +360,12 @@ export function installLlmReplay(ctx: Context, config: ReplayConfig): () => void
}
yield* replayEntry(entry, options.signal)
})()
})
}
const providers = config.providers ?? []
if (providers.length > 0) {
return ctx.llm.registerAdapter(providers.map(provider => provider.id), new ReplayAdapter(providers, replay))
}
return ctx.on('llm/stream', (options: GenerateOptions, _next) => replay(options))
}
export const name = 'llm-replay'
@@ -314,6 +383,8 @@ export interface Config {
* a nested-agent scenario; absent/empty for a single-session scenario.
*/
childFiles?: string[]
/** Optional replay-only provider catalog; absent or empty selects catch-all waterfall replay. */
providers?: ReplayProviderConfig[]
}
export function apply(ctx: Context, config: Config = {}): void {
@@ -329,5 +400,6 @@ export function apply(ctx: Context, config: Config = {}): void {
file,
...overrideFile !== undefined && overrideFile.length > 0 ? { overrideFile } : {},
...childFiles.length > 0 ? { childFiles } : {},
...config.providers !== undefined ? { providers: config.providers } : {},
})
}

View File

@@ -198,7 +198,7 @@ describe('loadReplayScript', () => {
})
})
describe('installLlmReplay (through the real waterfall)', () => {
describe('installLlmReplay (through the real LlmService)', () => {
function writeLog(...calls: StreamChunk[][]): void {
let seq = 1
const events: SessionEvent[] = []
@@ -217,6 +217,40 @@ describe('installLlmReplay (through the real waterfall)', () => {
expect(await drain(ctx.llm.stream({ provider: 'm', model: 'm', messages: [] }))).toEqual(TEXT_CHUNKS)
})
it('registers a replay-only provider catalog when configured', async () => {
writeLog(TEXT_CHUNKS)
const ctx = new Context()
await ctx.plugin(LlmService)
const dispose = installLlmReplay(ctx, {
file,
providers: [
{
id: 'deepseek',
name: 'DeepSeek',
models: [
{ id: 'flash' },
{ id: 'pro', name: 'Pro', description: 'Larger model' },
],
},
{ id: 'empty' },
],
})
expect(ctx.llm.listProviders()).toEqual([
{ id: 'deepseek', name: 'DeepSeek' },
{ id: 'empty', name: 'empty' },
])
await expect(ctx.llm.listModels('deepseek')).resolves.toEqual([
{ provider: 'deepseek', id: 'flash', name: 'flash' },
{ provider: 'deepseek', id: 'pro', name: 'Pro', description: 'Larger model' },
])
await expect(ctx.llm.listModels('empty')).resolves.toEqual([])
expect(await drain(ctx.llm.stream({ provider: 'deepseek', model: 'pro', messages: [] }))).toEqual(TEXT_CHUNKS)
dispose()
expect(ctx.llm.listProviders()).toEqual([])
})
it('serves the Nth call the Nth derived entry (positional)', async () => {
const second: StreamChunk[] = [
{ type: 'block-start', index: 0, blockType: 'text' },
@@ -574,11 +608,12 @@ describe('apply (the plugin entry)', () => {
expect(inject).toEqual(['llm'])
})
it('installs replay from an explicit config.file', async () => {
it('installs replay and its catalog from explicit config', async () => {
writeFileSync(file, sessionJsonl(TEXT_CHUNKS.map((c, i) => chunkEvent(i + 1, 1, 1, c))), 'utf8')
const ctx = new Context()
await ctx.plugin(LlmService)
apply(ctx, { file })
apply(ctx, { file, providers: [{ id: 'm', models: [{ id: 'm' }] }] })
expect(ctx.llm.listProviders()).toEqual([{ id: 'm', name: 'm' }])
expect(await drain(ctx.llm.stream({ provider: 'm', model: 'm', messages: [] }))).toEqual(TEXT_CHUNKS)
})