fix(hooks): run hooks in the session cwd; honest process-level config + best-effort session-start; surface systemMessage drop

Address review on the bridges:

- Hook cwd (blocking): the bridges never passed a workdir to runHook, so hooks
  ran in the executor default (the ACP server launch dir), not the session
  cwd — a hook doing `pwd`/relative reads/marker writes operated in the wrong
  tree. Both bridges now thread the agent's session `header.cwd` (the
  session/new.cwd) as the hook workdir for agent-scoped points. Regression per
  bridge: server cwd ≠ session cwd, a `pwd` hook proves it ran in the session
  workspace (proven red without the workdir).
- Example config honesty (blocking): `configPath: ./hooks.json` is read ONCE at
  load against the PROCESS cwd, not per-session — the comment/README now say so
  explicitly (a project-local per-session hooks.json is not discovered;
  TODO(per-session-hook-config)). The hooks-run-in-session-cwd fix above is the
  distinct, separately-documented half.
- Session-start timing (blocking): agent/session-start is a synchronous emit and
  the hook runs on a detached .then, so injected context is BEST-EFFORT — not
  guaranteed before the first request. Downgrade the contract in code comments +
  README + RFC (TODO(session-start-gating)) rather than implying "first request
  sees it", and add a no-wait regression that asserts the safe properties
  without pre-waiting for the inject.
- systemMessage (non-blocking): the merge collects merged.systemMessages but no
  bridge surfaced it. Warn per hook (like updatedInput) and document it as
  deferred in both READMEs + the RFC; tests assert the warn + non-surfacing.
This commit is contained in:
Tianyi Cui
2026-07-01 16:34:28 +08:00
parent f011699e43
commit 09c8e549b0
9 files changed
+189 -16

No files matched your search

@@ -39,13 +39,18 @@ Each bridge maps the neutral `MergedHookOutcome` from the shared lib onto the se
The config is parsed ONCE at load; a read/parse failure logs and registers nothing rather than crashing boot (a typo'd path must not take the agent down). Only `type: 'command'` hooks run — a `prompt`/`agent`/HTTP hook (CC) or an `async: true` / non-command hook (Codex) is parsed-and-skipped with a warning. The emit-listener paths (`session-start`, `subagent/start`) run detached, with their `inject` contained in a `.catch` that logs (a throwing inject must not break session boot or the loop).
### Where hooks run, and where their config comes from
Two different cwds, kept distinct on purpose. The hooks **themselves** run in the agent's **session workspace**: for the agent-scoped points the bridge threads the session's `cwd` (`session/new.cwd`, on the session header) to `runHook` as the process working directory, so a hook's `pwd` / relative-file read / marker write operates in the user's project tree, not the server's launch directory. The **config path**, by contrast, is **process-level**: `configPath` is resolved and parsed once at load against the process launch cwd, so a single `hooks.json` applies to the whole process — there is no per-session config discovery that reads a project-local `hooks.json` from each `session/new.cwd` (`TODO(per-session-hook-config)`). This is an honest limitation of the current cut: the example `cordis.yml` documents that its `./hooks.json` is process-level, not per-project.
## Deferred (faithful-but-degraded)
- **Tool-input rewrite.** A CC/Codex `updatedInput` is logged + warned, not honored — input rewrite is a deferred consistency-design problem ([the pre-tool-input-rewrite RFC](../../proposed/feature/2026-06-30-pre-tool-input-rewrite.md)), because the pre-execution args are read by `tool/call` audit + `assistant/message` history + ACP/tool-bash presentation, so an honest rewrite is a design unit, not a field.
- **Stop loop-guard** (`TODO(stop-loop-guard)`). CC/Codex break an infinite force-continue with `stop_hook_active` (true once a Stop hook fired this run) plus a max-consecutive cap; both are deferred. Today `stop_hook_active` is always `false`, so a Stop hook that unconditionally blocks would force-continue every step — a hook author must self-limit until the guard lands.
- **Permission `ask`** degrades to `deny` at the `tools/pre-execute` seam (`FIXME(permissions)` in the interception-seams RFC) — there is no interactive permission prompt yet.
- **Hook `continue:false` (hard halt).** A hook can ask to halt the whole run (CC/Codex `continue:false`); the shared merge folds it into `MergedHookOutcome.stop`/`stopReason`, but no bridge acts on it (`TODO(hook-continue-false)`) — the interception seams have no "hard-halt the agent" primitive yet (a Decision blocks/steers a single point, not the run). Deferred with the loop-guard work; the halt request is recorded in the `hook/result` log, and the hook keeps its per-point effect (decision/context) meanwhile.
- **Config discovery.** The path is explicit in `cordis.yml`; the full multi-layer CC/Codex precedence walk and the trust/hash model are not reimplemented (`TODO`).
- **Config discovery.** The path is explicit in `cordis.yml` and process-level (see above); the full multi-layer CC/Codex precedence walk, per-session project-local discovery, and the trust/hash model are not reimplemented (`TODO(per-session-hook-config)`).
- **Session-start / subagent-start context is best-effort, not gated (`TODO(session-start-gating)`).** `agent/session-start` is a synchronous emit and the bridge runs its hook on a detached `.then`, so the injected `additionalContext` is not guaranteed to land before the first turn reaches the model — a slow hook can miss the first request (the context then arrives as a later injection). `subagent/start` is sharper: an in-process provider may have already queued the child's prompt before the listener runs, and a short-lived child can finish before the detached inject fires. Making startup context a gated/awaited primitive is a loop-level change deferred to the interception seams; today the contract is "injected as soon as the hook resolves", not "before the first request". The bridge tests do NOT wait on the injection where they assert the guaranteed-timing behavior, so they document the real (best-effort) timing rather than masking it.
### Multiple hooks on one point run serially, not concurrently
+8 -5
View File
@@ -86,11 +86,14 @@
- id: tool-todo
name: '@deepseek-ai/dsh-tool-todo'
# The Claude Code hook bridge, pointed at a `hooks.json` in the session cwd. A
# scenario that ships `workspace/hooks.json` (copied into the cwd before the run)
# exercises the hooks path end-to-end; every other scenario has no such file, so
# the bridge's parse fails-soft and it registers nothing (a silent no-op — the
# ACP app loads no logger exporter, so the warning never reaches stdout).
# The Claude Code hook bridge. `configPath` is read ONCE at load and resolves
# `./hooks.json` against the PROCESS cwd (not per-session) — in these snapshot
# runs the harness launches the subprocess with process cwd = the scenario's temp
# workspace, so a scenario that ships `workspace/hooks.json` (copied into that cwd
# before the run) exercises the hooks path end-to-end; every other scenario has no
# such file, so the parse fails-soft and the bridge registers nothing (a silent
# no-op — the ACP app loads no logger exporter, so the warning never reaches
# stdout). Hooks themselves run in the session cwd (the bridge passes it as workdir).
- id: hooks-claude
name: '@deepseek-ai/dsh-hooks-claude'
config:
+10 -5
View File
@@ -96,11 +96,16 @@
- id: tool-todo
name: '@deepseek-ai/dsh-tool-todo'
# The Claude Code hook bridge, pointed at a `hooks.json` in the session cwd. With
# no such file present the parse fails-soft and the bridge registers nothing (a
# silent no-op); a session whose cwd holds a `hooks.json` runs those hooks on the
# interception seams. stdout is the ACP JSON-RPC channel — the bridge's warnings
# go through ctx.logger (no exporter here), never to stdout.
# The Claude Code hook bridge. `configPath` is PROCESS-LEVEL: it is read ONCE at
# load and the relative `./hooks.json` resolves against the ACP server's launch
# cwd, NOT each `session/new.cwd`. So a single `hooks.json` next to where the
# server starts applies to every session; a project-local, per-session hooks.json
# is NOT discovered (per-session config resolution is a TODO — see the bridge
# README). With no file present the parse fails-soft and the bridge registers
# nothing (a silent no-op). Hooks THEMSELVES run in the session cwd (the bridge
# passes it as the workdir); only WHERE the config is read from is process-level.
# stdout is the ACP JSON-RPC channel — the bridge's warnings go through ctx.logger
# (no exporter here), never to stdout.
- id: hooks-claude
name: '@deepseek-ai/dsh-hooks-claude'
config:
+4 -1
View File
@@ -25,7 +25,9 @@ In a `cordis.yml`:
projectDir: .
```
The config is parsed **once** at load. A read/parse failure is contained — the bridge logs a warning and registers nothing rather than crashing boot (a typo'd path must not take the agent down). Only `type: 'command'` hooks run; a `prompt`/`agent`/HTTP hook is parsed-and-skipped with a warning.
The config is parsed **once** at load. `configPath` is **process-level**: a relative path resolves against the process's launch cwd at load time, so a single config applies to the whole process — there is no per-session (`session/new.cwd`) config discovery yet (`TODO(per-session-hook-config)`). A read/parse failure is contained — the bridge logs a warning and registers nothing rather than crashing boot (a typo'd path must not take the agent down). Only `type: 'command'` hooks run; a `prompt`/`agent`/HTTP hook is parsed-and-skipped with a warning.
The hooks **themselves** run in the agent's session workspace: for the agent-scoped points the bridge passes the session's `cwd` (the `session/new.cwd`) as the hook process's working directory, so a hook's `pwd`/relative-path/marker operates in the user's project tree, not the server launch dir.
## Hook points → seam Decisions
@@ -48,4 +50,5 @@ Injected context carries an explicit `{ kind: 'plugin', plugin: 'hooks-claude' }
## Deferred (faithful-but-degraded)
- **`updatedInput` (tool-input rewrite)** is logged + warned, **not honored** — input rewrite is a deferred consistency-design problem ([the pre-tool-input-rewrite RFC](../../../docs/rfc/proposed/feature/2026-06-30-pre-tool-input-rewrite.md)).
- **`systemMessage`** (a hook's user-facing warning) is logged + warned, **not surfaced** — there is no user-message channel on these seams yet (only model-facing `additionalContext`). The shared merge collects it; the bridge does not yet render it.
- **Stop loop-guard.** CC breaks an infinite force-continue with `stop_hook_active` (true once a Stop hook has fired this run) plus a max-consecutive cap; both are deferred (`TODO(stop-loop-guard)`). Today `stop_hook_active` is always `false`, so a Stop hook that unconditionally blocks would force-continue every step — a hook author must self-limit until the guard lands.
+25 -2
View File
@@ -51,7 +51,13 @@ export const inject = ['bash']
/** Plugin config: where the CC hook config lives + substitution roots. */
export interface Config {
/** Path to a `hooks.json` or a settings file whose `hooks` key holds the config. */
/**
* Path to a `hooks.json` or a settings file whose `hooks` key holds the config.
* PROCESS-LEVEL: read once at load, a relative path resolves against the process
* launch cwd, so one config applies to the whole process.
* TODO(per-session-hook-config): per-session discovery of a project-local
* `hooks.json` from each `session/new.cwd` is not yet implemented.
*/
configPath: string
/** Replaces `${CLAUDE_PLUGIN_ROOT}` in command strings (the plugin's root dir). */
pluginRoot?: string
@@ -124,6 +130,12 @@ export function apply(ctx: Context, config: Config): void {
): Promise<MergedHookOutcome> {
const groups: MatcherGroup[] = parsed[point] ?? []
const outputs: HookOutput[] = []
// Run the hook in the AGENT'S session workspace (the `session/new` cwd on the
// session header), not the executor default (the ACP server's launch dir).
// A hook that does `pwd`, reads a relative file, or writes a marker must
// operate in the user's project tree. Absent for a no-agent run (falls back
// to the executor default).
const workdir = opts.agent?.session.header.cwd
for (const group of groups) {
if (!matchesMatcher(group.matcher, matchQuery, 'claude')) continue
for (const hook of group.hooks) {
@@ -138,6 +150,7 @@ export function apply(ctx: Context, config: Config): void {
const { output, durationMs } = await runHook(ctx.bash, hook, {
payload,
...hookEnv ? { env: hookEnv } : {},
...workdir !== undefined ? { cwd: workdir } : {},
...opts.signal ? { signal: opts.signal } : {},
defaultTimeoutMs,
trailingNewline: true,
@@ -149,6 +162,9 @@ export function apply(ctx: Context, config: Config): void {
if (output.updatedInput !== undefined) {
ctx.logger.warn(`hooks-claude: ${point} hook requested updatedInput, which is not yet honored (ignored)`)
}
if (output.systemMessage !== undefined) {
ctx.logger.warn(`hooks-claude: ${point} hook emitted a systemMessage, which is not yet surfaced (ignored)`)
}
if (session && opts.turn !== undefined) {
const stderrSummary = summarize(output.stderr)
appendHookResult(session, {
@@ -180,7 +196,14 @@ export function apply(ctx: Context, config: Config): void {
}
// --- SessionStart: emit (cannot block). Inject any additionalContext into the
// agent so the first request sees it. The matcher subject is the source. ---
// agent. The matcher subject is the source.
// TODO(session-start-gating): `agent/session-start` is a SYNCHRONOUS emit and
// this hook runs on a detached `.then`, so the injected context is BEST-EFFORT
// — it is not guaranteed to land before the first turn reaches the model. A
// slow hook can miss the first request (the context then arrives as a later
// injection turn). Gating startup on the hook is a loop-level change deferred
// to the interception seams; today the contract is "injected as soon as the
// hook resolves", not "before the first request". ---
ctx.on('agent/session-start', (agent, source) => {
void runPoint('SessionStart', source, sessionStartPayload(agent, source), { agent })
.then((merged) => {
@@ -454,3 +454,79 @@ describe('hooks-claude coverage — detached-listener catch handlers', () => {
expect(adapter.requests).toHaveLength(1) // loop survived the thrown inject
})
})
describe('hooks-claude coverage — hook runs in the session cwd, not the server cwd', () => {
it('runs an agent-scoped hook in the session workspace even when the executor default differs', async () => {
// The bug: the bridge passed no workdir, so hooks ran in the executor default
// (the server launch dir), not session/new.cwd. Here the executor default and
// the session cwd are DIFFERENT temp dirs; a PreToolUse hook writes `pwd` to a
// marker and we assert it ran in the SESSION cwd.
const serverDir = dir()
const sessionDir = dir()
const marker = join(sessionDir, 'where')
// The hook is invoked with cwd = session dir, so a relative marker path lands there.
hooks(serverDir, { PreToolUse: [{ hooks: [{ type: 'command', command: 'pwd > where' }] }] })
const adapter = new MockAdapter([toolCallResponse('c1', 'echo', {}), textResponse('done')])
const ctx = new Context()
await ctx.plugin(LlmService)
await ctx.plugin(SessionStore)
await ctx.plugin(SystemPrompt)
await ctx.plugin(ToolRegistry)
await ctx.plugin(AgentRegistry)
await ctx.plugin(AgentLoop, { agents: [] })
// Executor default cwd = serverDir (deliberately NOT the session cwd).
await ctx.plugin(LocalBashExecutor, { timeoutMs: 10_000, cwd: serverDir })
await ctx.plugin(HooksClaude, { configPath: join(serverDir, 'hooks.json') })
ctx.llm.registerAdapter(['mock'], adapter)
ctx.tools.register(defineTool({ name: 'echo', description: 'e', parameters: {}, async execute() { return [{ type: 'text', text: 'ok' }] } }))
const { SessionId } = await import('@deepseek-ai/dsh-session')
const handle = ctx.agents.create({ agentId: AgentId('a1'), sessionId: SessionId('s1'), meta: { cwd: sessionDir }, agentOptions: { model: 'mock' } })
handle.agent.send([{ type: 'text', text: 'go' }])
await waitForIdle(ctx, handle.agent as ReactLoopAgent)
expect(existsSync(marker)).toBe(true) // the marker landed in the SESSION dir
const { readFileSync } = await import('node:fs')
const where = readFileSync(marker, 'utf8').trim()
// `pwd` may resolve symlinks (/var → /private/var etc.), so compare basenames.
expect(where.endsWith(sessionDir.split('/').pop()!)).toBe(true)
await handle.dispose()
})
})
describe('hooks-claude coverage — systemMessage is warned, not surfaced', () => {
it('a hook emitting a systemMessage is logged as not-yet-surfaced', async () => {
const d = dir()
const s = sh(d, 'sm.sh', '#!/usr/bin/env bash\necho \'{"systemMessage":"heads up"}\'\n')
const path = hooks(d, { UserPromptSubmit: [{ hooks: [{ type: 'command', command: s }] }] })
const adapter = new MockAdapter([textResponse('ok')])
const ctx = await harness(path, adapter)
const warn = vi.fn(); ctx.logger.warn = warn as never
const agent = ctx.agentLoop.create(AgentId('a1'), { model: 'mock' })
agent.send([{ type: 'text', text: 'go' }])
await waitForIdle(ctx, agent)
expect(warn).toHaveBeenCalledWith(expect.stringContaining('systemMessage'))
// Not surfaced: the systemMessage text never reaches the model request.
expect(JSON.stringify(adapter.requests[0]!.messages)).not.toContain('heads up')
})
})
describe('hooks-claude coverage — SessionStart timing is best-effort (no-wait)', () => {
it('does NOT crash or block when the prompt is sent immediately (context is best-effort, may miss the first request)', async () => {
// Regression for the documented downgrade: session-start injection is
// detached, so a prompt sent immediately need not observe it. This asserts
// the SAFE properties (no crash, the turn still runs) WITHOUT waiting for the
// inject first — it documents the best-effort timing rather than masking it
// by pre-waiting for context/message (which the guaranteed-timing tests do).
const d = dir()
const s = sh(d, 'start.sh', '#!/usr/bin/env bash\necho \'{"hookSpecificOutput":{"hookEventName":"SessionStart","additionalContext":"late ctx"}}\'\n')
const path = hooks(d, { SessionStart: [{ hooks: [{ type: 'command', command: s }] }] })
const adapter = new MockAdapter([textResponse('ok')])
const ctx = await harness(path, adapter)
const agent = ctx.agentLoop.create(AgentId('a1'), { model: 'mock' })
// Send immediately — do NOT wait for the session-start inject.
agent.send([{ type: 'text', text: 'go' }])
await waitForIdle(ctx, agent)
expect(adapter.requests).toHaveLength(1) // the turn ran regardless of hook timing
})
})
+5 -1
View File
@@ -31,7 +31,9 @@ In a `cordis.yml`:
model: deepseek-v4
```
The config is parsed **once** at load; a read/parse failure is contained (logs + registers nothing). Only sync `type: 'command'` hooks run — a non-command or `async: true` hook is parsed-and-skipped with a warning. A hook accepts `timeout` or the `timeoutSec` alias. Events outside the five Codex points are dropped at parse.
The config is parsed **once** at load. `configPath` is **process-level** — a relative path resolves against the process launch cwd at load time, not per-session (`TODO(per-session-hook-config)`). A read/parse failure is contained (logs + registers nothing). Only sync `type: 'command'` hooks run — a non-command or `async: true` hook is parsed-and-skipped with a warning. A hook accepts `timeout` or the `timeoutSec` alias. Events outside the five Codex points are dropped at parse.
The hooks themselves run in the agent's session workspace: for the agent-scoped points the bridge passes the session's `cwd` as the hook process's working directory, so a hook operates in the user's project tree, not the server launch dir.
## Hook points → seam Decisions
@@ -52,3 +54,5 @@ Injected context carries an explicit `{ kind: 'plugin', plugin: 'hooks-codex' }`
## Deferred
**Stop loop-guard** (`TODO(stop-loop-guard)`): as in CC, a Stop hook that unconditionally blocks would force-continue every step (`stop_hook_active` is always `false` here); the loop-guard is deferred. A hook author must self-limit until it lands.
**`systemMessage`**: a hook's user-facing warning is logged + warned, not surfaced — there is no user-message channel on these seams yet (only model-facing `additionalContext`).
+18 -1
View File
@@ -38,7 +38,12 @@ export const inject = ['bash']
/** Plugin config: where the Codex hooks.json lives + the model name for payloads. */
export interface Config {
/** Path to a Codex `hooks.json`. */
/**
* Path to a Codex `hooks.json`. PROCESS-LEVEL: read once at load, a relative
* path resolves against the process launch cwd.
* TODO(per-session-hook-config): per-session project-local discovery from each
* `session/new.cwd` is not yet implemented.
*/
configPath: string
/** The model name stamped on every payload (Codex includes `model` on each event). */
model?: string
@@ -90,6 +95,10 @@ export function apply(ctx: Context, config: Config): void {
): Promise<MergedHookOutcome> {
const groups: MatcherGroup[] = parsed[point] ?? []
const outputs: HookOutput[] = []
// Run the hook in the agent's session workspace (the `session/new` cwd), not
// the executor default (the server launch dir) — a hook reading a relative
// file or `pwd` must see the user's project tree. Absent for a no-agent run.
const workdir = opts.agent?.session.header.cwd
for (const group of groups) {
// Codex matches with PURE regex (no literal fast path).
if (!matchesMatcher(group.matcher, matchQuery, 'codex')) continue
@@ -104,6 +113,7 @@ export function apply(ctx: Context, config: Config): void {
}
const { output, durationMs } = await runHook(ctx.bash, hook, {
payload,
...workdir !== undefined ? { cwd: workdir } : {},
...opts.signal ? { signal: opts.signal } : {},
defaultTimeoutMs,
trailingNewline: false, // Codex writes stdin WITHOUT a trailing newline.
@@ -126,6 +136,9 @@ export function apply(ctx: Context, config: Config): void {
output.additionalContext = output.stdout
}
outputs.push(output)
if (output.systemMessage !== undefined) {
ctx.logger.warn(`hooks-codex: ${point} hook emitted a systemMessage, which is not yet surfaced (ignored)`)
}
if (session && opts.turn !== undefined) {
const stderrSummary = summarize(output.stderr)
appendHookResult(session, {
@@ -154,6 +167,10 @@ export function apply(ctx: Context, config: Config): void {
}
// SessionStart: emit. Codex passes a plain-stdout hook's output as additionalContext.
// TODO(session-start-gating): a synchronous emit + detached `.then`, so the
// injected context is BEST-EFFORT — not guaranteed before the first turn reaches
// the model (a slow hook can miss the first request). Gating is a deferred
// loop-level change; the contract is "injected as soon as the hook resolves".
ctx.on('agent/session-start', (agent, source) => {
void runPoint('SessionStart', source, { ...base(agent, 'SessionStart', model), source }, { agent, plainStdoutAsContext: true })
.then((merged) => {
@@ -436,4 +436,41 @@ describe('hooks-codex coverage — decision mapping paths', () => {
expect(ran).toBe(false) // the matcher fired → the hook denied the tool
expect(events(agent).some(e => e.type === 'hook/invoked' && e.data.point === 'PreToolUse')).toBe(true)
})
it('a hook emitting a systemMessage is warned as not-yet-surfaced', async () => {
const d = dir()
hooks(d, { UserPromptSubmit: [{ hooks: [{ type: 'command', command: sh(d, 'sm.sh', '#!/usr/bin/env bash\necho \'{"systemMessage":"heads up"}\'\n') }] }] })
const adapter = new MockAdapter([textResponse('ok')])
const ctx = await harness(join(d, 'hooks.json'), adapter)
const warn = vi.fn(); ctx.logger.warn = warn as never
const agent = ctx.agentLoop.create(AgentId('a1'), { model: 'mock' })
agent.send([{ type: 'text', text: 'go' }]); await waitForIdle(ctx, agent)
expect(warn).toHaveBeenCalledWith(expect.stringContaining('systemMessage'))
expect(JSON.stringify(adapter.requests[0]!.messages)).not.toContain('heads up')
})
it('runs an agent-scoped hook in the session cwd, not the executor default', async () => {
// Same regression as the CC bridge: the Codex bridge must thread the session
// cwd as the hook workdir. Executor default = serverDir; session cwd =
// sessionDir; the PreToolUse hook's `pwd` marker must land in sessionDir.
const serverDir = dir()
const sessionDir = dir()
const marker = join(sessionDir, 'where')
hooks(serverDir, { PreToolUse: [{ hooks: [{ type: 'command', command: 'pwd > where' }] }] })
const adapter = new MockAdapter([toolCallResponse('c1', 'Bash', { command: 'x' }), textResponse('done')])
const ctx = new Context()
await ctx.plugin(LlmService); await ctx.plugin(SessionStore); await ctx.plugin(SystemPrompt)
await ctx.plugin(ToolRegistry); await ctx.plugin(AgentRegistry); await ctx.plugin(AgentLoop, { agents: [] })
await ctx.plugin(LocalBashExecutor, { timeoutMs: 10_000, cwd: serverDir })
await ctx.plugin(HooksCodex, { configPath: join(serverDir, 'hooks.json'), model: 'm' })
ctx.llm.registerAdapter(['mock'], adapter)
ctx.tools.register(defineTool({ name: 'Bash', description: 'b', parameters: { command: { type: 'string' } }, async execute() { return [{ type: 'text', text: 'ok' }] } }))
const { SessionId } = await import('@deepseek-ai/dsh-session')
const handle = ctx.agents.create({ agentId: AgentId('a1'), sessionId: SessionId('s1'), meta: { cwd: sessionDir }, agentOptions: { model: 'mock' } })
handle.agent.send([{ type: 'text', text: 'go' }])
await waitForIdle(ctx, handle.agent as ReactLoopAgent)
expect(existsSync(marker)).toBe(true)
expect(readFileSync(marker, 'utf8').trim().endsWith(sessionDir.split('/').pop()!)).toBe(true)
await handle.dispose()
})
})