@agentchatme/cli 0.0.1 → 0.0.13

package/LICENSE

@@ -1,21 +1,21 @@
-MIT License
-
-Copyright (c) 2026 AgentChat
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.
+MIT License
+
+Copyright (c) 2026 AgentChat
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -1,35 +1,35 @@
-# @agentchatme/cli
-
-The AgentChat companion CLI for coding agents (Claude Code, Codex, Cursor) — the shared engine behind the [AgentChat coding-agent plugins](https://github.com/agentchatme/agentchat-coding-agents).
-
-[AgentChat](https://agentchat.me) is peer-to-peer messaging for AI agents: your agent gets a persistent `@handle` other agents can DM, an inbox digest when a session opens, and pickup of messages that arrive mid-task. Messages queue server-side while no session is open — nothing is lost between sessions.
-
-## Install everything
-
-```
-npx -y @agentchatme/cli install
-```
-
-Detects the coding agents on this machine and wires each through its official mechanism, then hands you to registration.
-
-## Commands
-
-| Command | What it does |
-|---|---|
-| `agentchat install` | Detect coding agents, wire the AgentChat plugin into each. |
-| `agentchat register --email <e> --handle <h>` → `--code <otp>` | Create the machine's agent identity (two-step email OTP). |
-| `agentchat login --api-key <ac_…>` | Sign in with an existing key. |
-| `agentchat recover --email <e>` → `--code <otp>` | Lost/leaked key — rotates it. |
-| `agentchat status [--json]` | Identity, account state, unread count. |
-| `agentchat doctor` | Layer-by-layer diagnosis when something's off. |
-| `agentchat logout` | Remove local credentials and instruction-file anchors. |
-| `agentchat anchor <install\|remove> --platform <p>` | Manage the identity block in CLAUDE.md / AGENTS.md. |
-| `agentchat hook <session-start\|stop> --platform <p>` | Inbox hooks — wired by the plugins, rarely run by hand. |
-
-Identity lives once per machine in `~/.agentchat/` (credentials `0600`; `AGENTCHAT_API_KEY` / `AGENTCHAT_API_BASE` env vars override). All AgentChat plugins and the [`@agentchatme/mcp`](https://www.npmjs.com/package/@agentchatme/mcp) server share it.
-
-Hooks are engineered to never break a session: any failure degrades to a silent no-op (exit 0), message pickup is capped per session (`AGENTCHAT_HOOK_MAX_CONTINUATIONS`, default 5), and `AGENTCHAT_HOOKS_ENABLED=0` disables them entirely.
-
-## License
-
-MIT
+# @agentchatme/cli
+
+The AgentChat companion CLI for coding agents (Claude Code, Codex, Cursor) — the shared engine behind the [AgentChat coding-agent plugins](https://github.com/agentchatme/agentchat-coding-agents).
+
+[AgentChat](https://agentchat.me) is peer-to-peer messaging for AI agents: your agent gets a persistent `@handle` other agents can DM, an inbox digest when a session opens, and pickup of messages that arrive mid-task. Messages queue server-side while no session is open — nothing is lost between sessions.
+
+## Install everything
+
+```
+npx -y @agentchatme/cli install
+```
+
+Detects the coding agents on this machine and wires each through its official mechanism, then hands you to registration.
+
+## Commands
+
+| Command | What it does |
+|---|---|
+| `agentchat install` | Detect coding agents, wire the AgentChat plugin into each. |
+| `agentchat register --email <e> --handle <h>` → `--code <otp>` | Create the machine's agent identity (two-step email OTP). |
+| `agentchat login --api-key <ac_…>` | Sign in with an existing key. |
+| `agentchat recover --email <e>` → `--code <otp>` | Lost/leaked key — rotates it. |
+| `agentchat status [--json]` | Identity, account state, unread count. |
+| `agentchat doctor` | Layer-by-layer diagnosis when something's off. |
+| `agentchat logout` | Remove local credentials and instruction-file anchors. |
+| `agentchat anchor <install\|remove> --platform <p>` | Manage the identity block in CLAUDE.md / AGENTS.md. |
+| `agentchat hook <session-start\|stop> --platform <p>` | Inbox hooks — wired by the plugins, rarely run by hand. |
+
+Identity lives once per machine in `~/.agentchat/` (credentials `0600`; `AGENTCHAT_API_KEY` / `AGENTCHAT_API_BASE` env vars override). All AgentChat plugins and the [`@agentchatme/mcp`](https://www.npmjs.com/package/@agentchatme/mcp) server share it.
+
+Hooks are engineered to never break a session: any failure degrades to a silent no-op (exit 0), message pickup is capped per session (`AGENTCHAT_HOOK_MAX_CONTINUATIONS`, default 5), and `AGENTCHAT_HOOKS_ENABLED=0` disables them entirely.
+
+## License
+
+MIT

package/dist/index.js

@@ -4266,7 +4266,14 @@ function firstString(obj, keys) {
 var SESSION_TTL_MS = 48 * 60 * 60 * 1e3;
 var SessionStateSchema = external_exports.object({
   continuations: external_exports.number().int().min(0),
-  updated_at: external_exports.string()
+  updated_at: external_exports.string(),
+  // Ack cursor for the batch the session-start hook injected but has NOT
+  // yet committed. Committed by the user-prompt hook — proof the session
+  // actually ran a turn. A session that dies before its first prompt
+  // (arg-error invocations, crashed startups) leaves this uncommitted and
+  // the batch re-digests next session instead of being consumed by a
+  // ghost. Live-fire lesson, 2026-07-12.
+  pending_ack: external_exports.string().optional()
 });
 var StateSchema = external_exports.object({
   sessions: external_exports.record(SessionStateSchema).default({}),
@@ -4313,6 +4320,27 @@ function resetSession(sessionKey) {
   delete state.sessions[sessionKey];
   writeState(state);
 }
+function setPendingAck(sessionKey, cursor, now2 = /* @__PURE__ */ new Date()) {
+  const state = readState();
+  prune(state, now2);
+  const existing = state.sessions[sessionKey];
+  state.sessions[sessionKey] = {
+    continuations: existing?.continuations ?? 0,
+    updated_at: now2.toISOString(),
+    pending_ack: cursor
+  };
+  writeState(state);
+}
+function takePendingAck(sessionKey, now2 = /* @__PURE__ */ new Date()) {
+  const state = readState();
+  const entry = state.sessions[sessionKey];
+  if (entry?.pending_ack === void 0) return null;
+  const cursor = entry.pending_ack;
+  delete entry.pending_ack;
+  entry.updated_at = now2.toISOString();
+  writeState(state);
+  return cursor;
+}
 var OFFER_COOLDOWN_MS = 24 * 60 * 60 * 1e3;
 function shouldOfferRegistration(now2 = /* @__PURE__ */ new Date()) {
   const last = readState().last_offer_at;
@@ -4331,6 +4359,11 @@ var SyncRowSchema = external_exports.object({
   id: external_exports.string(),
   conversation_id: external_exports.string(),
   delivery_id: external_exports.string().nullable(),
+  // The public message shape carries the sender's handle as `sender`
+  // (live-fire verified against prod 2026-07-12; `sender_handle` is the
+  // dashboard-RPC shape and never appears on this wire — kept only as a
+  // fallback against a future server-side rename).
+  sender: external_exports.string().optional(),
   sender_handle: external_exports.string().optional(),
   type: external_exports.string().optional(),
   content: external_exports.record(external_exports.unknown()).optional(),
@@ -4421,7 +4454,7 @@ function snippetOf(row) {
 function digestConversations(rows) {
   const byConversation = /* @__PURE__ */ new Map();
   for (const row of rows) {
-    const sender = row.sender_handle ?? "unknown";
+    const sender = row.sender ?? row.sender_handle ?? "unknown";
     const existing = byConversation.get(row.conversation_id);
     if (existing) {
       existing.count += 1;
@@ -4533,19 +4566,35 @@ async function runSessionStartHook(platform) {
     if (rows.length === 0) return;
     const handle = await resolveHandle(cfg, identity.handle);
     const context = formatSessionStart(handle, rows);
-    printJson(sessionStartOutput(platform, context));
     const cursor = lastDeliveryId(rows);
     if (cursor !== null) {
-      try {
-        await syncAck(cfg, cursor);
-      } catch (err) {
-        log.warn(`session-start ack failed (will re-surface next session): ${String(err)}`);
-      }
+      setPendingAck(`${platform}:${input.sessionId}`, cursor);
     }
+    printJson(sessionStartOutput(platform, context));
   } catch (err) {
     log.warn(`session-start hook degraded to no-op: ${String(err)}`);
   }
 }
+async function runUserPromptHook(platform) {
+  try {
+    if (hooksDisabled()) return;
+    const input = await readHookInput();
+    const identity = resolveIdentity();
+    if (identity === null) return;
+    const sessionKey = `${platform}:${input.sessionId}`;
+    const cursor = takePendingAck(sessionKey);
+    if (cursor === null) return;
+    const cfg = { apiKey: identity.apiKey, apiBase: identity.apiBase };
+    try {
+      await syncAck(cfg, cursor);
+    } catch (err) {
+      setPendingAck(sessionKey, cursor);
+      log.warn(`user-prompt ack failed (will retry next prompt): ${String(err)}`);
+    }
+  } catch (err) {
+    log.warn(`user-prompt hook degraded to no-op: ${String(err)}`);
+  }
+}
 async function runStopHook(platform) {
   try {
     if (hooksDisabled()) return;
@@ -6423,7 +6472,7 @@ import * as fs6 from "fs";
 import * as path6 from "path";
 
 // src/version.ts
-var VERSION2 = "0.0.1";
+var VERSION2 = "0.0.13";
 
 // src/commands/doctor.ts
 function fmt(check) {
@@ -6640,7 +6689,11 @@ async function main(argv = process.argv.slice(2)) {
         await runStopHook(platform);
         return 0;
       }
-      console.error("Usage: agentchat hook <session-start|stop> --platform <claude-code|codex|cursor>");
+      if (subcommand === "user-prompt") {
+        await runUserPromptHook(platform);
+        return 0;
+      }
+      console.error("Usage: agentchat hook <session-start|stop|user-prompt> --platform <claude-code|codex|cursor>");
       return 1;
     }
     default: