docs(interactive): spec + plan for in-app interactive sessions
Replace the external wt.exe 'Run interactively' launch with an in-app streaming chat (persistent claude --input-format stream-json), rendered in the shared SessionTerminalView in task detail and Mission Control. Autonomous task execution is untouched. Mid-turn interrupt+redirect verified against CLI 2.1.191 via spike.
This commit is contained in:
Mika Kuns
101
docs/superpowers/plans/2026-06-26-in-app-interactive-sessions.md
Normal file
101
docs/superpowers/plans/2026-06-26-in-app-interactive-sessions.md
Normal file
@@ -0,0 +1,101 @@
|
|||||||
|
# Plan — In-App Interactive Sessions
|
||||||
|
|
||||||
|
Spec: `docs/superpowers/specs/2026-06-26-in-app-interactive-sessions-design.md`
|
||||||
|
|
||||||
|
Implement on the shared main tree. Commit explicit paths per task (never `git add -A`).
|
||||||
|
Build with `-c Release` (running Worker locks Debug). No real-Claude tests — fake the
|
||||||
|
process stream. Sonnet subagents. Autonomous `TaskRunner`/`ClaudeProcess` path stays untouched.
|
||||||
|
|
||||||
|
## Task 1 — StreamingClaudeSession (worker, new file)
|
||||||
|
- `Runner/StreamingClaudeSession.cs`: persistent `claude` process. Ctor takes resolved args,
|
||||||
|
working dir, seeded first prompt, a line callback, `WorkerConfig`. Reuse the
|
||||||
|
`ProcessStartInfo` shape + `MCP_TOOL_TIMEOUT="200000"` from `ClaudeProcess`.
|
||||||
|
- Keeps stdin open; sends the first prompt as a user-message JSON line (escape via
|
||||||
|
`JsonSerializer`).
|
||||||
|
- stdout/stderr read tasks → line callback; parse `result` events to track `IsTurnInFlight`.
|
||||||
|
- `SendUserMessageAsync(text, ct)` — enqueue/write a user-message JSON line; if
|
||||||
|
`IsTurnInFlight`, also `InterruptAsync`.
|
||||||
|
- `InterruptAsync(ct)` — write the control-protocol interrupt line; best-effort (swallow +
|
||||||
|
log on failure → queue fallback applies).
|
||||||
|
- `StopAsync` / `DisposeAsync` — close stdin, kill the tree, await exit.
|
||||||
|
- Injectable stream seam so a fake can drive it without a real `claude` binary.
|
||||||
|
- Test: `StreamingClaudeSessionTests` (fake stream) — first message emitted; `result` flips
|
||||||
|
`IsTurnInFlight` off; a sent message produces a second turn; mid-turn send calls interrupt
|
||||||
|
then delivers; interrupt throw → delivered at natural turn end; stop kills.
|
||||||
|
|
||||||
|
## Task 2 — LiveSessionRegistry (worker, new file)
|
||||||
|
- `Runner/LiveSessionRegistry.cs`: singleton; `Register(taskId, StreamingClaudeSession)`,
|
||||||
|
`bool TryGet(taskId, out session)`, `Unregister(taskId)`, `Task StopAsync(taskId)`.
|
||||||
|
- Test: register→get; unregister; second register stops+replaces; missing get returns false.
|
||||||
|
|
||||||
|
## Task 3 — InteractiveSessionService (worker, new file)
|
||||||
|
- `Planning/InteractiveSessionService.cs`: inject `IDbContextFactory`, `WorkerConfig`,
|
||||||
|
`ClaudeArgsBuilder` (or build args inline), `HubBroadcaster`, `LiveSessionRegistry`.
|
||||||
|
- `StartAsync(taskId, ct)`: resolve list working dir + seeded prompt (reuse the body of
|
||||||
|
`PlanningSessionManager.OpenInteractiveAsync` + `BuildInteractivePrompt`); build interactive
|
||||||
|
args (`--model PlanningAlias --permission-mode auto` + streaming flags); spawn the session
|
||||||
|
with a callback that does `HubBroadcaster.TaskMessage(taskId, "[stdout] " + line)`;
|
||||||
|
register; broadcast `InteractiveSessionStarted`. Reject if one is already live for the task.
|
||||||
|
- `SendAsync(taskId, text, ct)` → registry `TryGet` → `SendUserMessageAsync`.
|
||||||
|
- `StopAsync(taskId, ct)` → registry stop + `InteractiveSessionEnded`.
|
||||||
|
- Move `OpenInteractiveAsync`/`BuildInteractivePrompt` out of `PlanningSessionManager` if it
|
||||||
|
reads cleaner (or call into it). Remove the `InteractiveLaunchContext` terminal coupling.
|
||||||
|
- Test: `InteractiveSessionServiceTests` (fake session factory + fake broadcaster) — start
|
||||||
|
resolves dir, seeds prompt, registers, broadcasts started; missing working dir throws;
|
||||||
|
send routes; stop broadcasts ended.
|
||||||
|
|
||||||
|
## Task 4 — Remove terminal interactive path (worker)
|
||||||
|
- `Planning/Interfaces/ITerminalLauncher.cs` + `WindowsTerminalLauncher.cs`: delete
|
||||||
|
`LaunchInteractiveAsync`; remove `InteractiveLaunchContext` from `PlanningSessionContext.cs`.
|
||||||
|
Keep planning start/resume launches.
|
||||||
|
- Fix any references; ensure the planning launcher tests still build.
|
||||||
|
|
||||||
|
## Task 5 — Hub + Broadcaster + DI (worker)
|
||||||
|
- `Hub/WorkerHub.cs`: re-point `OpenInteractiveTerminalAsync` to
|
||||||
|
`InteractiveSessionService.StartAsync` (drop `_launcher.LaunchInteractiveAsync`); add
|
||||||
|
`Task SendInteractiveMessage(taskId, text)`, `Task StopInteractiveSession(taskId)`
|
||||||
|
(+ optional `InterruptInteractiveSession`).
|
||||||
|
- `Hub/HubBroadcaster.cs`: `InteractiveSessionStarted(taskId)`, `InteractiveSessionEnded(taskId)`.
|
||||||
|
- `Program.cs`: register `LiveSessionRegistry` + `InteractiveSessionService` singletons.
|
||||||
|
- Test: `WorkerHub` send routes to a fake service; start invokes the service.
|
||||||
|
|
||||||
|
## Task 6 — UI client + fakes (ui)
|
||||||
|
- `Services/Interfaces/IWorkerClient.cs` + `WorkerClient.cs`: `SendInteractiveMessageAsync(
|
||||||
|
taskId, text)`, `StopInteractiveSessionAsync(taskId)` (+ optional interrupt); events
|
||||||
|
`Action<string>? InteractiveSessionStartedEvent`, `InteractiveSessionEndedEvent` with
|
||||||
|
`On<...>` handlers. `OpenInteractiveTerminalAsync` keeps name/signature.
|
||||||
|
- Update hand-rolled `IWorkerClient` fakes in **both** Ui.Tests and Worker.Tests.
|
||||||
|
|
||||||
|
## Task 7 — StreamLineFormatter user bubble (ui)
|
||||||
|
- Render `type:"user"` NDJSON events as `LogKind.User` (add the kind if missing).
|
||||||
|
- Test: a `user` event yields a `LogKind.User` `LogLineViewModel` with the text.
|
||||||
|
|
||||||
|
## Task 8 — Shared composer state on the session VMs (ui, hot files)
|
||||||
|
- Add to `TaskMonitorViewModel` and `DetailsIslandViewModel` (factor a shared helper —
|
||||||
|
`InteractiveComposer` — to avoid duplication): `ComposerDraft`, `IsInteractiveLive`
|
||||||
|
(toggled by `InteractiveSessionStarted/Ended` for the subscribed task),
|
||||||
|
`SubmitComposerCommand` (CanExecute: non-empty draft && (`HasPendingQuestion` ||
|
||||||
|
`IsInteractiveLive`)). Route: pending question → existing `AnswerTaskQuestionAsync`; else →
|
||||||
|
`SendInteractiveMessageAsync`. Clear draft on submit; clear `IsInteractiveLive` on ended.
|
||||||
|
- `MissionControlViewModel`: `EnsureMonitor(taskId)` on `InteractiveSessionStarted`.
|
||||||
|
- Test: composer enabled while interactive-live; submit routes (chat vs answer) + clears;
|
||||||
|
ended clears live state.
|
||||||
|
|
||||||
|
## Task 9 — SessionTerminalView composer (ui)
|
||||||
|
- `Views/Islands/SessionTerminalView.axaml(.cs)`: optional composer docked bottom (styled
|
||||||
|
props `IsComposerVisible`, `ComposerText`, `SubmitCommand`, `ComposerPlaceholder`); TextBox
|
||||||
|
(Enter submits) + Send button. Reuse existing tokens (no inline values).
|
||||||
|
- Bind it in `MonitorPaneView.axaml` and `DetailsIslandView.axaml` to each VM's composer
|
||||||
|
state. Fold the existing AskUser banner into the composer's "answering" state if it reads
|
||||||
|
cleaner; otherwise leave the banner and add the composer below.
|
||||||
|
|
||||||
|
## Task 10 — Localization
|
||||||
|
- `en.json` + `de.json`: `interactive.composer.placeholder`, `.send`, `.stop`, plus any
|
||||||
|
"session ended" notice. Keep parity (Localization.Tests).
|
||||||
|
|
||||||
|
## Task 11 — Build + test + verify
|
||||||
|
- Build App + Worker `-c Release`; run Worker.Tests, Ui.Tests, Localization.Tests.
|
||||||
|
- Self-review diffs. **Manual smoke (real CLI) — flag to Mika:** (a) Run interactively opens
|
||||||
|
an in-app chat (no terminal) and streams; (b) sending a message mid-turn interrupts +
|
||||||
|
redirects; (c) stop kills the process; (d) session shows in both task detail and Mission
|
||||||
|
Control. Do not push.
|
||||||
@@ -0,0 +1,147 @@
|
|||||||
|
# In-App Interactive Sessions — Design
|
||||||
|
|
||||||
|
**Date:** 2026-06-26
|
||||||
|
**Status:** Proposed (awaiting approval)
|
||||||
|
|
||||||
|
## Goal
|
||||||
|
|
||||||
|
Replace the external Windows-Terminal "Run interactively" session with an **in-app
|
||||||
|
streaming chat**, rendered in the existing `SessionTerminalView` in **both task detail and
|
||||||
|
Mission Control**. Keep everything inside the app — no `wt.exe` pop-out. Autonomous task
|
||||||
|
execution is **untouched** (stays one-shot, non-interactive).
|
||||||
|
|
||||||
|
## Decisions (brainstorm)
|
||||||
|
|
||||||
|
1. **Engine: persistent streaming session.** One `claude` process kept alive with
|
||||||
|
`--input-format stream-json`; user messages pushed over stdin.
|
||||||
|
2. **Scope: interactive sessions only.** The autonomous `TaskRunner`/`ClaudeProcess` run
|
||||||
|
loop, review, queue, and worktree machinery are NOT changed.
|
||||||
|
3. **Placement: shared `SessionTerminalView`** — the in-app session + composer appear in the
|
||||||
|
task-detail session surface and in the Mission Control monitor pane.
|
||||||
|
4. **Full replace.** "Run interactively" now opens the in-app session; the
|
||||||
|
`WindowsTerminalLauncher.LaunchInteractiveAsync` path is removed. **Planning** sessions
|
||||||
|
keep using `wt` (untouched).
|
||||||
|
5. **Send semantics: interrupt + redirect** mid-turn (control protocol), with automatic
|
||||||
|
*queue-for-next-turn* fallback if interrupt is unavailable.
|
||||||
|
|
||||||
|
## What an interactive session is (unchanged semantics, new transport)
|
||||||
|
|
||||||
|
Today (`PlanningSessionManager.OpenInteractiveAsync` + `WindowsTerminalLauncher`):
|
||||||
|
`claude --model <PlanningAlias> --permission-mode auto "<task title+description>"` in the
|
||||||
|
**list's working dir**, env `MAX_THINKING_TOKENS=20000`, full default toolset, relies on the
|
||||||
|
globally-registered `claudedo` MCP. **Ephemeral** — no worktree, no `task_run` record, no
|
||||||
|
status change, no review.
|
||||||
|
|
||||||
|
We keep all of that. Only the transport changes: instead of a `wt` window, the same
|
||||||
|
`claude` invocation runs as a persistent stream-json process owned by the worker, its output
|
||||||
|
streamed into the app and its stdin fed from an in-app composer.
|
||||||
|
|
||||||
|
> Honest tradeoff: the `wt` terminal gave the full Claude Code TUI (slash-command UX,
|
||||||
|
> interactive prompts). An in-app stream-json chat is plainer — type messages, watch streamed
|
||||||
|
> output. `--permission-mode auto` means no blocking permission prompts (so headless works),
|
||||||
|
> but it is a simpler surface than the real TUI. Accepted per the "full replace" decision.
|
||||||
|
|
||||||
|
## The streaming engine
|
||||||
|
|
||||||
|
Flags: `--model <PlanningAlias> --permission-mode auto --input-format stream-json
|
||||||
|
--output-format stream-json --verbose --replay-user-messages` in the list working dir, env
|
||||||
|
`MAX_THINKING_TOKENS=20000`. No `--mcp-config`/`--allowedTools` (interactive uses the global
|
||||||
|
MCP + default tools, exactly as today).
|
||||||
|
|
||||||
|
- First stdin message = the seeded interactive prompt:
|
||||||
|
`{"type":"user","message":{"role":"user","content":[{"type":"text","text":"…"}]},"parent_tool_use_id":null}\n`
|
||||||
|
(stdin stays open).
|
||||||
|
- A stdout read task forwards each NDJSON line to a callback (→ broadcast + the session's log)
|
||||||
|
and detects `result` events (turn boundary; the process then idles for the next message).
|
||||||
|
- `SendUserMessageAsync(text)` writes a user-message JSON line; if a turn is in flight, also
|
||||||
|
`InterruptAsync()` (control-protocol interrupt) so Claude pivots immediately. If interrupt
|
||||||
|
is unavailable, the message lands when the current turn ends → automatic queue fallback.
|
||||||
|
- **Interrupt is verified working** (spike, 2026-06-26, CLI 2.1.191). Exact shape:
|
||||||
|
`{"type":"control_request","request_id":"<id>","request":{"subtype":"interrupt"}}` — no
|
||||||
|
`initialize` handshake needed; `control_response {"subtype":"success"}` confirms
|
||||||
|
synchronously; the same process then accepts the redirect and runs a fresh turn with
|
||||||
|
context intact.
|
||||||
|
- **Interrupt artifact:** the aborted turn emits a `result` with `is_error=true,
|
||||||
|
subtype="error_during_execution"`. The session must treat an interrupt-induced result as
|
||||||
|
*"turn aborted, continue"* (drain the queued redirect), **not** as a session failure.
|
||||||
|
Tolerate the incidental `system:init`/`system:status`/`rate_limit_event`/hook events that
|
||||||
|
also appear in the stream.
|
||||||
|
- `--replay-user-messages` echoes each sent message back on stdout as a `user` event, so it
|
||||||
|
rides the existing stream pipeline into the timeline (ordered + confirmed) with no extra
|
||||||
|
broadcast surface.
|
||||||
|
- The session ends only when the **user stops it** (kill the process tree) — an interactive
|
||||||
|
session has no auto-finalize and never enters review. No queue slot is involved (it is
|
||||||
|
launched directly, not via the autonomous picker).
|
||||||
|
|
||||||
|
## Surface changes
|
||||||
|
|
||||||
|
**Worker**
|
||||||
|
- `Runner/StreamingClaudeSession.cs` (new) — persistent process + send/interrupt/stop; reuse
|
||||||
|
the `ProcessStartInfo` shape + `MCP_TOOL_TIMEOUT` from `ClaudeProcess`; streams via a line
|
||||||
|
callback; `IsTurnInFlight`. Cancellation kills the tree.
|
||||||
|
- `Runner/LiveSessionRegistry.cs` (new, singleton) — `taskId → StreamingClaudeSession`
|
||||||
|
(`Register`/`TryGet`/`Unregister`/`Stop`), mirrors `PendingQuestionRegistry`.
|
||||||
|
- `Planning/InteractiveSessionService.cs` (new) — owns interactive lifecycle: `StartAsync(
|
||||||
|
taskId)` resolves the list working dir + seeded prompt (reuse `OpenInteractiveAsync`'s
|
||||||
|
body), spawns the session, registers it, wires output to `HubBroadcaster.TaskMessage`,
|
||||||
|
broadcasts `InteractiveSessionStarted`; `SendAsync(taskId, text)`; `StopAsync(taskId)` →
|
||||||
|
`InteractiveSessionEnded`.
|
||||||
|
- `Planning/WindowsTerminalLauncher.cs` + `Planning/Interfaces/ITerminalLauncher.cs` — remove
|
||||||
|
`LaunchInteractiveAsync` (+ `InteractiveLaunchContext`). Planning start/resume stay.
|
||||||
|
- `Hub/WorkerHub.cs` — `OpenInteractiveTerminalAsync` re-pointed to
|
||||||
|
`InteractiveSessionService.StartAsync` (no terminal); add `SendInteractiveMessage(taskId,
|
||||||
|
text)`, `StopInteractiveSession(taskId)` (+ optional `InterruptInteractiveSession`).
|
||||||
|
- `Hub/HubBroadcaster.cs` — `InteractiveSessionStarted(taskId)`,
|
||||||
|
`InteractiveSessionEnded(taskId)`. Log lines reuse the existing `TaskMessage(taskId, line)`.
|
||||||
|
- `Program.cs` — register `LiveSessionRegistry` + `InteractiveSessionService`.
|
||||||
|
|
||||||
|
**UI**
|
||||||
|
- `Views/Islands/SessionTerminalView.axaml(.cs)` — add an optional composer (styled
|
||||||
|
properties: `IsComposerVisible`, `ComposerText`, `SubmitCommand`, `ComposerPlaceholder`).
|
||||||
|
Both hosts (task detail + Mission Control) get it by binding their VM's composer state.
|
||||||
|
- `StreamLineFormatter` — render `type:"user"` NDJSON events as a `LogKind.User` bubble.
|
||||||
|
- A small shared composer concept on `TaskMonitorViewModel` **and** `DetailsIslandViewModel`
|
||||||
|
(factor a helper to avoid duplication): `ComposerDraft`, `SubmitComposerCommand`,
|
||||||
|
`IsInteractiveLive` (set by `InteractiveSessionStarted/Ended`). Submit →
|
||||||
|
`SendInteractiveMessageAsync`; clear draft. (If a pending AskUser question exists, the same
|
||||||
|
composer answers it — keep the existing answer route.)
|
||||||
|
- `MissionControlViewModel` — `EnsureMonitor(taskId)` on `InteractiveSessionStarted` so the
|
||||||
|
session appears as a monitor; mark it interactive.
|
||||||
|
- `Services/Interfaces/IWorkerClient.cs` + `WorkerClient.cs` — `SendInteractiveMessageAsync`,
|
||||||
|
`StopInteractiveSessionAsync` (+ optional interrupt); events
|
||||||
|
`InteractiveSessionStartedEvent`/`InteractiveSessionEndedEvent`. `OpenInteractiveTerminalAsync`
|
||||||
|
keeps its name/signature (now starts the in-app session). Update hand-rolled fakes in **both**
|
||||||
|
test projects (`iworkerclient_fakes_sync`).
|
||||||
|
- `TasksIslandViewModel.RunInteractivelyAsync` — unchanged call site; now opens/focuses the
|
||||||
|
in-app session surface instead of a terminal.
|
||||||
|
- Localization `interactive.*` / `missionControl.chat.*` (en/de, parity enforced).
|
||||||
|
|
||||||
|
**Tests**
|
||||||
|
- `StreamingClaudeSessionTests` (fake process stream, no real Claude): first message streams;
|
||||||
|
`result` idles; a sent message starts another turn; mid-turn send calls `InterruptAsync`
|
||||||
|
then delivers; interrupt-failure degrades to queue; stop kills.
|
||||||
|
- `LiveSessionRegistryTests` — register/get/unregister/stop.
|
||||||
|
- `InteractiveSessionServiceTests` — start resolves working dir + seeds prompt + registers +
|
||||||
|
broadcasts started; send routes to the session; stop broadcasts ended (fake session +
|
||||||
|
broadcaster).
|
||||||
|
- `TaskMonitorViewModelTests` / `DetailsIslandViewModelTests` — composer enabled while
|
||||||
|
interactive-live; submit invokes client + clears; `user` line renders; question route still
|
||||||
|
answers.
|
||||||
|
|
||||||
|
## Risks / open questions
|
||||||
|
|
||||||
|
- **Interrupt protocol shape — RESOLVED** (spike 2026-06-26, see "The streaming engine").
|
||||||
|
Mid-turn interrupt works on CLI 2.1.191 with the documented shape; the queue fallback is a
|
||||||
|
genuine fallback now, not the expected path. Re-verify if the CLI version changes.
|
||||||
|
- **Plainer than the TUI** — slash-command/interactive-prompt UX differs (accepted).
|
||||||
|
- **Auto-mode editing the list working dir directly** (no worktree) — this is the *existing*
|
||||||
|
interactive behavior, unchanged here.
|
||||||
|
- **No real-Claude tests** (project rule) — the live loop is covered only by the fake stream;
|
||||||
|
real interrupt/redirect is a **manual verification gap** to flag.
|
||||||
|
|
||||||
|
## Non-goals
|
||||||
|
|
||||||
|
- Changing autonomous task execution / review / queue / worktrees.
|
||||||
|
- Interactive sessions producing run records, worktrees, or review (stays ephemeral).
|
||||||
|
- Worktree isolation for interactive edits; image/attachment messages in the composer.
|
||||||
|
- Removing planning's `wt` terminal launch.
|
||||||
Reference in New Issue
Block a user