@moreai/code 0.1.0-beta.1 → 0.1.0-beta.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (7) hide show
  1. package/README.md +190 -275
  2. package/dist/daemon.mjs +1195 -1190
  3. package/dist/moreai.mjs +204 -204
  4. package/dist/resources/file-acl/macos/moreai-file-acl +0 -0
  5. package/media/helpers.webp +0 -0
  6. package/media/session.webp +0 -0
  7. package/package.json +14 -8
package/README.md CHANGED
@@ -1,337 +1,252 @@
1
- # More AI Code (`moreai`)
1
+ <h1 align="center">More AI Code</h1>
2
2
 
3
- A terminal coding agent that runs on its own. The desktop app is optional: the CLI
4
- starts a bundled local engine when no More AI app is running.
3
+ <p align="center">
4
+ <b>An AI coding agent for your terminal.</b><br>
5
+ Your own model keys, an engine on your machine, and cloud runs when you want them.
6
+ </p>
5
7
 
6
- ## Install the preview
8
+ <p align="center">
9
+ <a href="https://www.npmjs.com/package/@moreai/code"><img alt="npm" src="https://img.shields.io/npm/v/@moreai/code/beta?label=npm%20beta&color=ff7a45"></a>
10
+ <img alt="Node.js 22+" src="https://img.shields.io/badge/node-%E2%89%A522-339933">
11
+ <img alt="macOS · Linux · Windows" src="https://img.shields.io/badge/macOS%20%C2%B7%20Linux%20%C2%B7%20Windows-555">
12
+ <a href="https://more-ai.net"><img alt="more-ai.net" src="https://img.shields.io/badge/more--ai.net-website-111"></a>
13
+ </p>
7
14
 
8
- Requires Node.js 22 or newer on macOS, Linux or Windows.
15
+ <p align="center">
16
+ <img alt="More AI Code exploring a repository, editing a file, writing a test, running it and summarising the change" src="https://cdn.jsdelivr.net/npm/@moreai/code@0.1.0-beta.2/media/session.webp" width="860">
17
+ </p>
18
+
19
+ ## Install
9
20
 
10
21
  ```sh
11
- npm install -g @moreai/code@beta
12
- moreai setup
22
+ npm install -g @moreai/code
13
23
  cd your-project
14
24
  moreai
15
25
  ```
16
26
 
17
- The `beta` tag follows preview releases. To check for a new version and update
18
- the global installation:
27
+ Requires **Node.js 22 or newer** on macOS, Linux or Windows. The first launch
28
+ helps you connect a model; nothing else is required — no account, no desktop app.
29
+
30
+ ## What it does
31
+
32
+ - **Works in your repository.** Reads and searches the code, edits files and
33
+ shows each change as a diff, runs commands and tests, keeps a checklist of
34
+ its own plan, and answers in plain language.
35
+ - **Runs on your models.** About 40 providers: Anthropic, OpenAI, Google
36
+ Gemini, OpenRouter, DeepSeek, Groq, local models through Ollama or LM Studio,
37
+ any OpenAI-compatible endpoint, and ChatGPT or GitHub Copilot subscriptions.
38
+ Keys stay on your machine.
39
+ - **Asks before it acts.** Five access modes, from read-only planning to full
40
+ autonomy. Commands run in an OS sandbox where one is available, and risky
41
+ ones are held for you.
42
+ - **Splits big questions across helpers.** Parallel helper agents research
43
+ different parts of the codebase at once and report back.
44
+ - **Moves work to the cloud.** With a More AI account, a task can keep running
45
+ on our servers with your laptop closed and come back as a branch or a pull
46
+ request.
47
+ - **Listens.** `Ctrl+T` dictates into the prompt.
48
+ - **Speaks your language.** The interface is translated into 35 languages.
49
+
50
+ ## Getting started
19
51
 
20
52
  ```sh
21
- moreai update
22
- moreai update --install
23
- moreai --version
53
+ moreai # interactive session in this folder
54
+ moreai setup # guided setup: model, optional account and services
55
+ moreai auth # connect or change an AI provider
56
+ moreai doctor # check that everything is ready
24
57
  ```
25
58
 
26
- `moreai update` checks the npm release tag and shows the available version;
27
- `--install` runs the global npm update. You can also update manually with
28
- `npm install -g @moreai/code@beta`. The CLI checks for updates in the
29
- interactive session and shows a quiet notice when a new version is available.
30
-
31
- If you received a packaged `.tgz` archive instead of installing from npm,
32
- install it with `npm install -g ./moreai-code-<version>.tgz`, then run the same
33
- `moreai setup` step. The archive includes the headless engine. Tarball installs
34
- do not receive updates until you install a newer archive or switch to the npm
35
- `beta` tag.
36
-
37
- `moreai setup` walks through a More AI account and an AI provider:
38
-
39
- - **AI provider** gives the agent a model on this computer. Connect an API key,
40
- a local model (Ollama, LM Studio) or a supported provider subscription
41
- through `moreai auth`. If a key is already in your environment
42
- (`ANTHROPIC_API_KEY`, `OPENAI_API_KEY`, …) the first run offers it at the top
43
- of the list, and `moreai auth env` connects every one it finds. A More AI
44
- account is not required to use your own provider.
45
- - **More AI account** is optional and adds what needs a server: cloud
46
- sessions that keep working with this computer off (`moreai cloud`), session
47
- and settings sync, More AI dictation, and Google services (Gmail, Calendar,
48
- Drive) for the agent. Sign in with your email; approve the request on
49
- another signed-in device or enter the emailed six-digit code. The account
50
- session stays in the local engine, never in your shell history.
51
- - **External services** give the agent tools from a connector, such as an MCP
52
- server. `moreai connectors` lists the catalog whenever you want one.
53
-
54
- You can run these steps separately at any time:
59
+ If a provider key is already in your environment (`ANTHROPIC_API_KEY`,
60
+ `OPENAI_API_KEY`, …), the first launch offers it at the top of the list, and
61
+ `moreai auth env` connects every one it finds.
55
62
 
56
- ```sh
57
- moreai login # sign in to More AI
58
- moreai account # show the signed-in account and plan
59
- moreai logout # sign out; keeps your own provider connections
60
- moreai auth # connect or change an AI provider
61
- moreai auth env # connect the provider keys found in the environment
62
- moreai auth status # list connected providers
63
- moreai connectors # choose an external service
64
- moreai connectors list # list connected services
65
- moreai doctor # check the local engine and setup
66
- ```
63
+ In a session, type a task and press **Enter**. While the agent works you can
64
+ keep typing to steer it, **Esc** interrupts, and **Shift+Tab** changes the
65
+ access mode. `/help` lists every command.
67
66
 
68
- When a provider signs in through the browser and you are on a remote machine
69
- over SSH, the sign-in page cannot reach back to the terminal on its own. The
70
- wizard prints the `ssh -L` command that forwards the sign-in port, or you can
71
- paste an API key instead.
72
-
73
- ### External services from the terminal
74
-
75
- `moreai connectors` (or `/connectors` in a session) is one searchable list:
76
- what is connected — Google services from your More AI account (Gmail,
77
- Calendar, Meet, Tasks, Drive) and MCP servers alike, each with its state —
78
- then the whole catalog by category. Type to search, **Enter** connects (or
79
- fixes a connection that needs a new sign-in), **Space** turns a connected
80
- service on or off, **Ctrl+D** disconnects it (press twice). Google services
81
- belong to your More AI account, so a service connected on the desktop, the web
82
- or the phone is already on here. Choosing one while signed out explains why
83
- the account is needed and opens the sign-in right there; once you are in, the
84
- connection continues on its own. A service whose sign-in is kept with the
85
- account can also be connected on this computer only, without one. For a script or SSH
86
- session, use the commands directly:
67
+ ## Access modes
87
68
 
88
- ```sh
89
- moreai connectors catalog # show service names and IDs
90
- moreai connectors add github # connect a listed service
91
- moreai connectors add https://mcp.example.com/mcp # connect a custom MCP URL
92
- moreai connectors list # inspect connection IDs and status
93
- moreai connectors off ID # keep a service but take its tools away
94
- moreai connectors on ID # turn it back on
95
- moreai connectors remove ID # disconnect by the ID shown in the list
96
- ```
69
+ | Mode | What the agent may do |
70
+ | --- | --- |
71
+ | `plan` | Read the code and propose a plan for you to approve; then it builds |
72
+ | `read-only` | Read and answer; never change anything |
73
+ | `accept-edits` | Edit files on its own; ask before running commands |
74
+ | `auto` *(default)* | Work on its own; anything it cannot show to be safe is declined |
75
+ | `full` | Run everything except a small set of destructive commands |
97
76
 
98
- For a service that needs a token, pipe it to `--token-stdin` so the secret does
99
- not appear in the command arguments or shell history:
77
+ Pick one with **Shift+Tab** in a session or `--mode <mode>` on the command line.
100
78
 
101
- ```sh
102
- printf '%s\n' "$MCP_TOKEN" | moreai connectors add github --token-stdin
103
- ```
79
+ ## Helper agents
104
80
 
105
- Some services open a browser to finish sign-in. `moreai connectors list` shows
106
- whether each connection is ready. `moreai auth` remains the command for the
107
- AI model that answers your questions.
81
+ <p align="center">
82
+ <img alt="Three parallel helper agents researching webhooks, each with its result, then the combined answer" src="https://cdn.jsdelivr.net/npm/@moreai/code@0.1.0-beta.2/media/helpers.webp" width="860">
83
+ </p>
108
84
 
109
- ### Sign in without an interactive terminal
85
+ For broad questions the agent can start several **helper agents** in
86
+ parallel. Each gets its own line — what it was asked, what it is doing now,
87
+ and when it finishes, its tool count, tokens and time with the first line of
88
+ its report. The run's token total shows how much of it went to helpers.
89
+ Helpers only read; the main agent makes the changes.
110
90
 
111
- Starting sign-in from a script or SSH session requires an email argument:
91
+ Start a task with `@explore`, `@scout` or `@general` to hand it to a helper
92
+ directly:
112
93
 
113
94
  ```sh
114
- moreai login you@example.com
95
+ moreai "@explore how does the billing webhook verify signatures?"
115
96
  ```
116
97
 
117
- If it sends an email code, pipe the **latest** six-digit code to the verification
118
- command. Piping keeps the code out of process arguments and shell history:
119
-
120
- ```sh
121
- printf '%s\n' "$MOREAI_EMAIL_CODE" | moreai login you@example.com --code-stdin
122
- ```
98
+ ## Cloud sessions
123
99
 
124
- If it returns a push approval challenge, the command prints its ID. Approve the
125
- matching confirmation code on another device, then run:
100
+ With a More AI account (`moreai login`), a task can run in the cloud instead
101
+ of on this computer. You can follow it here, or answer it from the web app or
102
+ your phone.
126
103
 
127
104
  ```sh
128
- moreai login --challenge <id> --wait
105
+ moreai cloud run "add tests for the parser" # move this folder into a cloud session
106
+ moreai --cloud "add tests for the parser" # the same
107
+ moreai cloud # your cloud sessions and their state
108
+ moreai cloud open <id> # the conversation and its link
109
+ moreai cloud finish <id> # bring the work back again
129
110
  ```
130
111
 
131
- To use an email code instead, run `moreai login --challenge <id> --fallback`,
132
- then pipe the code as above. No account token is printed by these commands.
112
+ `moreai cloud run` uploads the folder as it is — your branch, commits and
113
+ uncommitted changes, never files that look like secrets or are over 25 MB —
114
+ and follows the task in the terminal. **Ctrl+C** stops following; the task
115
+ keeps running. When it finishes, the work comes back as a **new branch**
116
+ `more-ai/<topic>-<id>`, and your own branch and edits are never touched. In a
117
+ GitHub or GitLab repository it is pushed and a pull request is opened with
118
+ your own `gh` / `glab` login; `--finish push|branch|folder` chooses otherwise.
119
+ Inside a session, `/cloud` does the same.
133
120
 
134
- ## Work in a project
121
+ ## Scripts and pipes
135
122
 
136
123
  ```sh
137
- moreai # interactive session in the current folder
138
- moreai "fix the failing test" # open a session and run a task
139
- moreai -c # continue the latest session for this folder
140
- moreai sessions # list this folder's sessions
141
- moreai -p "add types" # run once, print the result, exit
124
+ moreai -p "summarise the open TODOs" # run once, print the answer, exit
142
125
  git diff | moreai -p "review this diff"
143
- moreai -p --json "…" # one JSON event per line
126
+ moreai -p --json "add types to utils.ts" # one JSON event per line
127
+ moreai -c # continue the newest session here
128
+ moreai sessions # this folder's sessions
144
129
  ```
145
130
 
146
131
  | Option | Meaning |
147
132
  | --- | --- |
148
- | `-p, --print` | Run once without the interactive UI |
149
- | `--json` | With `-p`, output machine-readable event lines |
133
+ | `-p, --print` | Run once without the interactive interface |
134
+ | `--json` | With `-p`: machine-readable events, one per line |
150
135
  | `-c, --continue` | Continue this folder's newest session |
151
136
  | `--session <id>` | Resume a specific session |
152
- | `-m, --model <id>` | Choose a model for this run |
153
- | `--provider <id>` | Disambiguate the provider for a model |
154
- | `--mode <mode>` | `plan`, `read-only`, `accept-edits`, `auto`, or `full` |
155
- | `-C, --cwd <dir>` | Set the working folder |
156
- | `--cloud` | Run the task in a cloud session instead (see below) |
137
+ | `-m, --model <id>` | Choose the model for this run |
138
+ | `--provider <id>` | Say which provider serves `--model` |
139
+ | `--mode <mode>` | Access mode (see above) |
140
+ | `-C, --cwd <dir>` | Work in another folder |
141
+ | `--cloud` | Run the task in a cloud session |
157
142
  | `--finish <how>` | With `--cloud`: `pr`, `push`, `branch` or `folder` |
158
- | `--detach` | With `--cloud`: start it and return without following |
143
+ | `--detach` | With `--cloud`: start it and return immediately |
144
+ | `--lang <id>` | Interface language for this run |
145
+
146
+ ## In the session
159
147
 
160
- In the session, press **Enter** to send, **Ctrl+J** (or a trailing `\`) for a
161
- newline, **Esc** to interrupt, and **Shift+Tab** to change access mode. Use
162
- `/help` to see the available commands. You can type while the agent is working
163
- to steer its run. The status line shows the model, how much of its context
164
- window is left (yellow under 25 %, red under 10 %), the branch and the access
165
- mode.
148
+ | Key | Action |
149
+ | --- | --- |
150
+ | Enter | Send (or steer the running task) |
151
+ | Ctrl+J, or `\` at line end | New line |
152
+ | Esc | Interrupt the run; twice clears the draft |
153
+ | Shift+Tab | Change the access mode |
154
+ | Ctrl+T | Dictate (Enter or Ctrl+T to insert, Esc to discard) |
155
+ | ↑ ↓ | Move between lines, then through your history |
156
+ | Ctrl+O | Expand a large paste back into editable text |
157
+ | Ctrl+D, or Ctrl+C twice | Leave; the command to resume is printed |
158
+
159
+ Useful commands: `/model` (pick a model; **Ctrl+F** pins it), `/sessions`,
160
+ `/connectors` (Gmail, Calendar, Drive, GitHub and other services for the
161
+ agent), `/cloud`, `/language`, `/login`, `/status`, `/help`.
166
162
 
167
- While the agent works, what it reads and searches folds into one
168
- **Explored** line, edits show as diffs, and its checklist stays above the input.
163
+ ## Voice input
169
164
 
170
- ### Helper agents
165
+ **Ctrl+T** (or `/voice`) records from the microphone and puts the text at the
166
+ cursor without sending it. Transcription uses More AI dictation once you sign
167
+ in, your own transcription service, or an on-device model set up in the More
168
+ AI desktop app. Recording uses the ffmpeg that comes with the desktop app, or
169
+ `ffmpeg`, `sox`, `parecord` or `arecord` from your `PATH`.
171
170
 
172
- The agent can split broad research across parallel **helper agents**. Each one
173
- gets its own line under the step that started it — what it was asked, what it
174
- is doing now, and when it finishes, its tool count, tokens and time with the
175
- first line of its report. The run's token total says how much of it the helpers
176
- spent. Helpers only read; the main agent makes the changes.
171
+ ## Account and services
177
172
 
178
- Start a task with `@explore`, `@scout` or `@general` to hand it to that helper
179
- directly, even when automatic delegation is off:
173
+ A More AI account is optional. It adds cloud sessions, sync of your sessions to
174
+ the web and phone apps, More AI dictation, and Google services (Gmail,
175
+ Calendar, Drive) for the agent.
180
176
 
181
177
  ```sh
182
- moreai "@explore how does the billing webhook verify signatures?"
178
+ moreai login # sign in with your email (a code or an approval on another device)
179
+ moreai account # the signed-in account and plan
180
+ moreai connectors # connect a service; `moreai connectors list` shows them
181
+ moreai logout # sign out; your own provider keys stay
183
182
  ```
184
183
 
185
- ### Voice input
186
-
187
- **Ctrl+T** (or `/voice`) records from the microphone: a red dot, a timer and a
188
- level meter show it is listening. **Enter** or **Ctrl+T** stops and puts the
189
- text at the cursor — it is not sent, so you can read it first — and **Esc**
190
- throws the recording away. A recording stops by itself after two minutes.
191
-
192
- Transcription uses whatever the More AI app's dictation is set to: More AI
193
- dictation once you `/login`, your own transcription service, or an on-device
194
- model (Settings → Dictation in the app). The audio is recorded by this terminal,
195
- so the first time the system asks whether your terminal app may use the
196
- microphone. If a transcription fails, the recording is kept: **Ctrl+T** tries
197
- again.
198
-
199
- To record, `moreai` uses the ffmpeg that comes with the More AI app, or one of
200
- `ffmpeg`, `sox`, `parecord`, `arecord` on your PATH. `MOREAI_VOICE_RECORDER`
201
- names any other command that prints raw 16-bit 16 kHz mono audio on stdout
202
- (arguments separated by spaces).
203
-
204
- ### Leaving a session
205
-
206
- **Ctrl+D** on an empty draft, **Ctrl+C** twice or `/quit` ends the session and
207
- hands the terminal back: the screen is cleared and one line remains with the
208
- command that resumes the session. Only the visible screen is cleared — what you
209
- ran before `moreai` stays in the scrollback — and the terminal gets its own
210
- title back. `MOREAI_KEEP_SCREEN=1` leaves the session on screen instead. A
211
- crash never clears the screen, so the error stays readable.
212
-
213
- ### Choosing a model
214
-
215
- `/model` (or `/models`) opens the model picker: the current model, the models
216
- you pinned (shared with the desktop app's picker), your recent picks, then every
217
- connected provider — a large catalog such as OpenRouter grouped by vendor, with
218
- the context window and price where the provider reports them.
219
-
220
- - **Type** to search model names, providers and vendors (`sonnet`,
221
- `or claude` for OpenRouter's Claude models, `anthropic`).
222
- - **↑ ↓** move, **PgUp PgDn** page, **← →** switch the provider tab.
223
- - **Enter** uses the model for this session and remembers it as your latest
224
- pick, so a new session — here or in the desktop app — starts on it.
225
- - **Ctrl+F** pins or unpins a model; **Ctrl+A** connects another provider;
226
- **Esc** clears the search, then closes.
227
-
228
- `/model <text>` switches immediately when the text names exactly one model
229
- (`/model glm-4.5`, `/model openrouter/anthropic/claude-sonnet-4.6`); anything
230
- else opens the picker with that search.
231
-
232
- ### Editing and pasting
233
-
234
- The input box is a multi-line editor: **↑ ↓** move between lines (and reach your
235
- history from the first or last line of an empty or recalled draft),
236
- **Alt+← →** move by word, **Ctrl+A / Ctrl+E** jump to the line start or end,
237
- **Ctrl+U / Ctrl+K / Ctrl+W** delete. A long draft scrolls inside the box.
238
-
239
- A large paste (more than 8 lines or 800 characters) becomes a single
240
- `[Pasted text #1 · 200 lines]` item: **Backspace** removes it whole,
241
- **Ctrl+O** turns it back into editable text, and the agent receives the full
242
- text when you send. **Ctrl+C** clears the draft and keeps it in history; **Esc**
243
- twice clears it too.
244
-
245
- ### Language
246
-
247
- The terminal agent speaks the same 35 languages as the desktop app and picks
248
- one without asking: `--lang` or `MOREAI_LANG` first, then your `/language`
249
- choice, then the desktop app's language, then the terminal's locale
250
- (`LC_ALL` / `LANG`) and finally the system language. `/language` lists them all
251
- (type to search; **Automatic** follows the app and the system), and
252
- `/language de` switches straight away. Slash commands, flags and key names stay
253
- in English everywhere, and `--json` output is always English.
184
+ From a script or an SSH session: `moreai login you@example.com`, then pipe the
185
+ emailed code to `moreai login you@example.com --code-stdin`. When a provider
186
+ signs in through the browser on a remote machine, the CLI prints the
187
+ `ssh -L` command that forwards the sign-in back to your terminal.
254
188
 
255
- ```sh
256
- moreai --lang ja # this run only
257
- MOREAI_LANG=ru moreai # the same, from the environment
258
- ```
189
+ ## How it runs
259
190
 
260
- ## Cloud sessions
191
+ `moreai` starts a small background engine on your machine the first time it
192
+ needs one; later sessions, the More AI desktop app and the VS Code extension
193
+ share it, so a session started in one place continues in another. If the
194
+ desktop app is running with **Settings → Agent tools → External clients** on,
195
+ the CLI uses the app's engine instead.
196
+
197
+ Provider keys are encrypted with a key kept in the system keyring (macOS
198
+ Keychain, Windows Credential Manager, or the Secret Service on Linux). A Linux
199
+ server without a keyring service stores them in a file only your user can read
200
+ (`~/.config/More AI/secrets.json`), the way most command-line tools keep tokens.
201
+
202
+ ## Privacy
261
203
 
262
- With a More AI account, a task can run in the cloud instead of on this
263
- computer — it keeps going when you close the laptop, and you can follow it or
264
- answer it from the web app or your phone.
204
+ Your code and keys stay on your machine unless you move a task to the cloud.
205
+ More AI collects usage analytics to find bugs; with AI traces on (the default)
206
+ the content of runs is included so failed runs can be debugged. API keys are
207
+ never sent. The first interactive session asks once; change it any time:
265
208
 
266
209
  ```sh
267
- moreai cloud run "add tests for the parser" # move this folder into a cloud session
268
- moreai --cloud "add tests for the parser" # the same
269
- moreai cloud # your cloud sessions and their state
270
- moreai cloud open <id> # the conversation and its link
271
- moreai cloud finish <id> # bring the work back again
272
- moreai cloud cancel <id> # stop the running task
273
- moreai cloud models # models a cloud run can use
210
+ moreai analytics off # nothing is sent
211
+ moreai analytics usage # usage only, no conversation content
212
+ moreai analytics on # usage and AI traces
274
213
  ```
275
214
 
276
- `moreai cloud run` packages the folder as it is — your branch, commits and
277
- uncommitted changes, never files that look like secrets or are larger than
278
- 25 MB — uploads it and starts the task, then follows it in the terminal: the
279
- current step, the answer when it is done. **Ctrl+C** stops following; the task
280
- keeps running (`--detach` returns straight away).
281
-
282
- When the task finishes, the work comes back on its own as a **new branch**
283
- `more-ai/<topic>-<id>`; your own branch and uncommitted edits are never
284
- touched. In a repository on GitHub or GitLab it is then pushed and a pull
285
- request is opened with your own `gh` / `glab` login; `--finish` chooses
286
- otherwise: `push`, `branch` (local only) or `folder` (copy the changed files
287
- into this folder, keeping your edits). The cloud runs on your plan's models,
288
- or on your own provider key if you uploaded it in the More AI app
289
- (Settings → Privacy); `-m` picks one of `moreai cloud models`.
290
-
291
- Inside a session, `/cloud` lists the cloud sessions (Enter shows one) and
292
- `/cloud <task>` moves the task there.
293
-
294
- ## Engine and data
295
-
296
- `moreai doctor` and `/status` name the backend this build talks to. Installed
297
- packages use production (`api.more-ai.net`); only `MOREAI_ENV=dev`, at build or
298
- run time, points the CLI and its engine at the test tier, and the session's
299
- welcome screen then shows a `DEV` badge.
300
-
301
-
302
- Without the desktop app, `moreai` launches a bundled headless engine on this
303
- machine. Provider credentials are encrypted with a key kept in the operating
304
- system's keyring (macOS Keychain, Windows Credential Manager, or the Secret
305
- Service on Linux — GNOME Keyring or KWallet). A Linux server with no keyring
306
- service stores them unencrypted in a file only your user can read
307
- (`~/.config/More AI/secrets.json`), the way most command-line tools keep
308
- tokens. The CLI and VS Code can connect to the same running engine; sessions
309
- and settings persist across CLI restarts.
310
-
311
- If the desktop app is running with **External clients** enabled, the CLI attaches
312
- to its engine, so sessions appear on both surfaces. If the app is running with
313
- External clients disabled, enable it in **Settings → Agent tools**, or close the
314
- app so the CLI can start its own engine. `moreai doctor` reports this condition.
315
-
316
- ## Usage analytics
317
-
318
- More AI collects usage analytics to find bugs and see which features work:
319
- commands you use, errors, and how agent runs go. With AI traces on (the
320
- default), the content of runs is included too, so failed runs can be debugged.
321
- API keys are never sent. The first interactive session asks once; you can
322
- change it at any time:
215
+ `DO_NOT_TRACK=1` or `MOREAI_TELEMETRY=0` turns analytics off for any process
216
+ that sees it.
217
+
218
+ ## Update and uninstall
323
219
 
324
220
  ```sh
325
- moreai analytics # show the current setting
326
- moreai analytics usage # analytics without conversation content
327
- moreai analytics off # turn analytics off
328
- moreai analytics on # analytics and AI traces
221
+ moreai update # is there a newer version?
222
+ moreai update --install # install it
223
+ npm uninstall -g @moreai/code
329
224
  ```
330
225
 
331
- Inside a session, `/analytics` opens the same choice. `DO_NOT_TRACK=1` or
332
- `MOREAI_TELEMETRY=0` turns analytics off for any process that sees it. The
333
- setting belongs to this machine's engine, so it is shared with the More AI
334
- desktop app (Settings → Privacy) when both are installed.
226
+ The background engine notices an update or uninstall and stops by itself once
227
+ nothing is using it, so the next session runs the new version.
228
+
229
+ Version 0.1.0-beta.1 cannot see newer versions: update it once with
230
+ `npm install -g @moreai/code@latest`. Its engine predates the self-stop and
231
+ keeps running until you restart the computer or run
232
+ `pkill -f @moreai/code/dist/daemon.mjs` (macOS, Linux).
233
+
234
+ ## Troubleshooting
235
+
236
+ - **`moreai doctor`** checks the engine, the connection and your setup, and says
237
+ what to fix.
238
+ - **"No AI provider is connected"**: run `moreai auth`, or `moreai auth env` if
239
+ your key is in an environment variable.
240
+ - **The desktop app is running but the CLI cannot attach**: turn on
241
+ **Settings → Agent tools → External clients** in the app, or close the app.
242
+ - **Commands are held on a Linux server**: in `auto` the agent runs commands
243
+ on its own only inside a sandbox (bubblewrap on Linux). Without one, switch to
244
+ `accept-edits` (**Shift+Tab**) to approve each command, or use `--mode full`
245
+ in an environment you trust.
246
+
247
+ Questions and bug reports: [support@more-ai.net](mailto:support@more-ai.net).
248
+
249
+ ## License
335
250
 
336
- © 2026 More AI. All rights reserved. Proprietary software — see `LICENSE`;
337
- the open-source components it bundles are listed in `THIRD_PARTY_LICENSES.txt`.
251
+ © 2026 More AI. All rights reserved. Proprietary software — see `LICENSE`.
252
+ The open-source components it bundles are listed in `THIRD_PARTY_LICENSES.txt`.