aiterm-mcp 0.29.20 → 0.29.21

package/CHANGELOG.md

@@ -7,6 +7,12 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.29.21] - 2026-09-02
+
+### Fixed
+
+- Detect Cursor prompts left in the composer when the arrow and first prompt line share one row, and fail dispatch instead of returning a successful receipt for text that was not submitted.
+
 ## [0.29.20] - 2026-09-01
 
 ### Changed
@@ -1469,7 +1475,8 @@ prototype (preserved under `prototype/python/` as the porting source and referen
   `ubuntu-latest` for Node 18/20/22, publishing to npm on `v*` tags with
   provenance.
 
-[Unreleased]: https://github.com/kitepon/aiterm-mcp/compare/v0.29.20...HEAD
+[Unreleased]: https://github.com/kitepon/aiterm-mcp/compare/v0.29.21...HEAD
+[0.29.21]: https://github.com/kitepon/aiterm-mcp/compare/v0.29.20...v0.29.21
 [0.29.20]: https://github.com/kitepon/aiterm-mcp/compare/v0.29.19...v0.29.20
 [0.29.19]: https://github.com/kitepon/aiterm-mcp/compare/v0.29.18...v0.29.19
 [0.29.18]: https://github.com/kitepon/aiterm-mcp/compare/v0.29.17...v0.29.18

package/README.ja.md

@@ -153,7 +153,7 @@ runtime-error store は canonical dotagents config の `collection.enabled: true
 場合だけ収集し、既定OFF、network送信は行いません。tag起点CIのnpm provenance（OIDC Trusted
 Publishing）で公開し、GitHub Release が Official MCP Registry を再登録します。
 
-**状態:** 開発継続中 · 現行公開版 **v0.29.20** · 動作対象は Linux · WSL2 · macOS · Windows ネイティブ · MIT · [変更履歴](CHANGELOG.md)。
+**状態:** 開発継続中 · 現行公開版 **v0.29.21** · 動作対象は Linux · WSL2 · macOS · Windows ネイティブ · MIT · [変更履歴](CHANGELOG.md)。
 
 ### 更新と巻き戻し
 
@@ -187,7 +187,7 @@ pty_read(id, { wait: true })       → 削減済みの出力を読む（完了
 
 同じprimitiveが別エージェントのTUIを宿す。`agent_launch`の`harness`はagent loop・認証・hook・session・transcriptを所有する実行基盤、`model`は独立した選択。起動processは直接CLIと同じproject/user環境を使い、通常config、MCP、plugin、skill、permission、trust、memory、historyをcopy・filter・置換しない。
 
-`aiterm.agent-launch-result.v1`は正規`harness`を返し、旧`provider`は互換fieldとして残す。同じ`harness`はagent dispatch、`aiterm-wait`、`agent_configure`、`pty_list`のagent行にも載り、旧vendor／provider／agent fieldは互換用に残る。Codexは通常rollout、Grok CLIは通常session event、Claudeはlaunch固有Stop hook、Cursorは通常agent transcript末尾の`turn_ended`を完了正本に使う。`pty_send`は非ブロックdispatchで、vendor別完了境界を表すopaqueな整数`event_cursor`を返し、完了通知は`aiterm-wait`を親のターンを塞がない別processで受ける。
+`aiterm.agent-launch-result.v1`は正規`harness`を返し、旧`provider`は互換fieldとして残す。同じ`harness`はagent dispatch、`aiterm-wait`、`agent_configure`、`pty_list`のagent行にも載り、旧vendor／provider／agent fieldは互換用に残る。Codexは通常rollout、Grok CLIは通常session event、Claudeはlaunch固有Stop hook、Cursorは通常agent transcript末尾の`turn_ended`を完了正本に使う。`pty_send`は非ブロックdispatchで、vendor別完了境界を表すopaqueな整数`event_cursor`を返し、完了通知は`aiterm-wait`を親のターンを塞がない別processで受ける。送信textがCursorのcomposerへ残った場合は成功receiptを返さず失敗する。
 
 `agent_launch`は任意の`write_scope`も受ける。Codex／Grokのread-onlyは`--sandbox read-only`、Cursorは公式`--mode ask`で実効化する。path説明は同等CLI引数がないためdeclaration-only。
 

package/README.md

@@ -169,7 +169,7 @@ collection is off by default and performs no network I/O. It ships via
 tag-triggered CI with npm provenance (OIDC Trusted Publishing); the GitHub
 Release re-registers the Official MCP Registry entry.
 
-**Status:** actively maintained · current public release **v0.29.20** · runs on Linux · WSL2 · macOS · native Windows (tmux on POSIX, the tmux-CLI-compatible [psmux](https://github.com/psmux/psmux) on native Windows — no WSL required) · MIT · see the [CHANGELOG](CHANGELOG.md).
+**Status:** actively maintained · current public release **v0.29.21** · runs on Linux · WSL2 · macOS · native Windows (tmux on POSIX, the tmux-CLI-compatible [psmux](https://github.com/psmux/psmux) on native Windows — no WSL required) · MIT · see the [CHANGELOG](CHANGELOG.md).
 
 ### Update and rollback
 
@@ -209,7 +209,7 @@ pty_read(id, { wait: true })       → read the token-reduced output, completion
 
 The same primitive hosts another agent's TUI. `agent_launch` starts a selected execution harness inside a fresh persistent terminal and returns a `session_id`. `harness` names the component that owns the agent loop, authentication, hooks, session, and transcript; `model` remains an independent choice. The launched process sees the same project and user environment as a direct CLI invocation: normal configuration, MCPs, plugins, skills, permissions, trust decisions, memory, and history are not copied, filtered, or replaced. Aiterm adds only completion correlation and a non-user sub-agent context containing `role=subagent`, the parent session, delegation depth, lineage, and `delegation_allowed=true`.
 
-The human-readable launch text is accompanied by an `aiterm.agent-launch-result.v1` structured receipt containing the canonical `harness`; the old `provider` field remains for compatibility. The same `harness` is carried by agent dispatch, `aiterm-wait`, `agent_configure`, and agent rows in `pty_list`, while their old vendor/provider/agent fields remain compatibility fields. Codex completion comes from its normal durable rollout transcript, Grok CLI from its normal session events, Claude Code from a launch-specific Stop hook settings addition, and Cursor from its normal agent transcript's terminal `turn_ended` record. Sending to any agent session is a non-blocking **dispatch** — the call returns immediately with an opaque, harness-specific integer `event_cursor`, and completion arrives via [`aiterm-wait`](#completion-push-for-parent-agents-aiterm-wait).
+The human-readable launch text is accompanied by an `aiterm.agent-launch-result.v1` structured receipt containing the canonical `harness`; the old `provider` field remains for compatibility. The same `harness` is carried by agent dispatch, `aiterm-wait`, `agent_configure`, and agent rows in `pty_list`, while their old vendor/provider/agent fields remain compatibility fields. Codex completion comes from its normal durable rollout transcript, Grok CLI from its normal session events, Claude Code from a launch-specific Stop hook settings addition, and Cursor from its normal agent transcript's terminal `turn_ended` record. Sending to any agent session is a non-blocking **dispatch** — the call returns immediately with an opaque, harness-specific integer `event_cursor`, and completion arrives via [`aiterm-wait`](#completion-push-for-parent-agents-aiterm-wait). If submitted text remains in Cursor's composer, its dispatch fails instead of returning a successful receipt.
 
 `agent_launch` accepts an optional `write_scope`: either `"read-only"` or a human-readable description of writable paths. Codex/Grok use `--sandbox read-only`; Cursor uses its official read-only `--mode ask`. A path description remains declaration-only because these CLI launch surfaces provide no equivalent path allowlist flag.
 

package/dist/core.js

@@ -19,7 +19,7 @@ import { sleep, currentUid, runtimeStateBase, safeStatSize, readFileRange, write
 import { GROK_MODEL_DEFAULTS, realGrokHome, resolveAndValidateGrokAuth, assertGrokModelAvailable, grokEventsTranscript, latestGrokCompletion, observeGrokDone, buildGrokAgentCmd, grokLaunchNote, grokEnvTokens, grokTuiReady, GROK_COMPOSER_MARKER_RE, grokFooterHasConfiguration, grokTranscriptText, createGrokAgentMetadata, } from "./harnesses/grok.js";
 import { bindCodexTranscriptSession, latestCodexCompletion, observeCodexDone, buildCodexAgentCmd, codexLaunchNote, codexTuiReady, codexLaunchBlockingDialog, CODEX_COMPOSER_MARKER_RE, codexModelChoice, codexEffortChoice, codexMoreReasoningChoice, codexTranscriptText, createCodexAgentMetadata, } from "./harnesses/codex.js";
 import { OPERATION_ID_RE, CLAUDE_RESULT_MAX_BYTES, CLAUDE_EFFORTS, agentManagedClaudeSettingsPath, agentClaudeResultPath, agentClaudeOperationPath, agentClaudeApprovalReceiptPath, agentClaudeDispatchReceiptPath, validateOperationId, readClaudeResultText, assertClaudeAuthenticationReady, buildClaudeAgentCmd, claudeLaunchNote, claudeTuiReady, CLAUDE_COMPOSER_MARKER_RE, createClaudeAgentMetadata, } from "./harnesses/claude.js";
-import { bindCursorTranscriptSession, cursorTurnBoundary, latestCursorCompletion, observeCursorDone, cursorTranscriptText, assertCursorAuthenticationReady, assertCursorModelAvailable, buildCursorAgentCmd, cursorPromptWithLineage, createCursorAgentMetadata, cursorLaunchNote, cursorEffortNavigation, cursorTuiReady, CURSOR_COMPOSER_MARKER_RE, validateCursorModelEffort, } from "./harnesses/cursor.js";
+import { bindCursorTranscriptSession, cursorTurnBoundary, latestCursorCompletion, observeCursorDone, cursorTranscriptText, assertCursorAuthenticationReady, assertCursorModelAvailable, buildCursorAgentCmd, cursorPromptWithLineage, createCursorAgentMetadata, cursorLaunchNote, cursorEffortNavigation, cursorTuiReady, CURSOR_COMPOSER_CONTENT_MARKER_RE, validateCursorModelEffort, } from "./harnesses/cursor.js";
 import { resolveAgentBin, resolveThroughlineBin, runThroughlineHandoffContext, isWindowsNativeExecutable, agentBinForPaneShell, resolveWinPaneShell } from "./agent-resolver.js";
 export { AitermError } from "./errors.js";
 export { tmuxSpawnEnv } from "./tmux-runtime.js";
@@ -2417,7 +2417,7 @@ function agentSubmitResidueOnScreen(kind, screen, tail) {
         : kind === "claude"
             ? CLAUDE_COMPOSER_MARKER_RE
             : kind === "cursor"
-                ? CURSOR_COMPOSER_MARKER_RE
+                ? CURSOR_COMPOSER_CONTENT_MARKER_RE
                 : GROK_COMPOSER_MARKER_RE;
     let markerIdx = -1;
     for (let i = lines.length - 1; i >= 0; i--) {
@@ -2464,6 +2464,15 @@ export function agentSubmitResidueWarning(name, residue) {
         `pty_read(${name}, screen:true) で状態を確認してから、座礁していれば pty_key(${name}, "Enter") で再 submit、` +
         `破棄するなら pty_key(${name}, "Escape") を使う。盲目的に Enter を送らない（queued だった場合の二重 submit 防止）。`);
 }
+function assertAgentSubmitDelivered(name, kind, residue) {
+    if (kind !== "cursor" || residue.residue !== true)
+        return;
+    throw new AitermError(`submit_residue=true vendor=${kind} session=${name}\n` +
+        "送信 text がcomposerへ残っており、turnを開始できませんでした。", 2);
+}
+export function __testAssertAgentSubmitDelivered(name, kind, residue) {
+    assertAgentSubmitDelivered(name, kind, residue);
+}
 async function settleAgentDoneScreenImpl(sample, sleepFn, opts = {}) {
     const minDelayMs = opts.minDelayMs ?? AGENT_DONE_SETTLE_MIN_MS;
     const pollMs = opts.pollMs ?? AGENT_DONE_SCREEN_SETTLE_POLL_MS;
@@ -2592,6 +2601,13 @@ export async function sendInitialAgentPrompt(name, text, o = {}) {
                 `起動時promptがCursorのcomposerへ残り、Enter再送後もturnを開始できませんでした。`, 2);
         }
     }
+    try {
+        assertAgentSubmitDelivered(name, meta.kind, residue);
+    }
+    catch (e) {
+        setInitialPromptState(meta, "failed");
+        throw e;
+    }
     return {
         text: `initial_prompt=pending vendor=${meta.kind} event_cursor=${startOffset} harness=${agentHarness(meta.kind)}\n` +
             `起動時 prompt を送信した。${agentDispatchGuide(name, startOffset)}` +
@@ -2850,6 +2866,7 @@ export async function dispatchAgentTurn(name, text, o = {}) {
     await sleep(AGENT_SUBMIT_DELAY_MS);
     sendKey(name, "Enter", { preserveAgentOperation: meta.kind === "claude" });
     const residue = await detectAgentSubmitResidue(name, meta.kind, dispatchText);
+    assertAgentSubmitDelivered(name, meta.kind, residue);
     return {
         schema: "aiterm.agent-dispatch.v1",
         session_id: meta.aiterm_session,

package/dist/harnesses/cursor.js

@@ -392,6 +392,9 @@ export function cursorLaunchNote(model, effort, meta) {
     return `${modelNote}${writeScopeLaunchNote("cursor", meta?.write_scope)}`;
 }
 export const CURSOR_COMPOSER_MARKER_RE = /(?:^|\n)\s*(?:>|→|->)\s*(?:\n|$)/m;
+// 送信後の残留検出では、本文が矢印と同じ行に残る現行UIもcomposerとして扱う。
+// ready判定には使わない。本文入りcomposerを入力待ちと誤認させないため。
+export const CURSOR_COMPOSER_CONTENT_MARKER_RE = /^\s*(?:->|>|→)(?:\s|$)/;
 const CURSOR_FOLLOWUP_MARKER_RE = /(?:^|\n)\s*(?:→|->)\s*Add a follow-up\b/im;
 const CURSOR_START_PROMPT_MARKER_RE = /(?:^|\n)\s*(?:→|->)\s*Plan,\s*search,\s*build anything\s*(?:\n|$)/im;
 export function cursorTuiReady(screen) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "aiterm-mcp",
-  "version": "0.29.20",
+  "version": "0.29.21",
   "mcpName": "io.github.kitepon/aiterm-mcp",
   "description": "Persistent terminal MCP with one harness-based launcher for Claude Code, Codex CLI, Grok CLI, and Cursor Agent CLI, plus durable PTYs for SSH, containers, and REPLs.",
   "keywords": [