1<img src="internal/assets/zot-logo.png" alt="zot" width="130" height="130" />
2
3# zot
4
5Yet another coding agent harness, lightweight and written (vibe-slopped) in go.
6
7- one static binary.
8- built-in providers for anthropic, openai/codex, kimi, deepseek, google gemini, and ollama/openai-compatible local models.
9- four tools (read, write, edit, bash).
10- three run modes (interactive tui, print, json).
11- built-in telegram bot.
12- extensions in any language via subprocess + json-rpc. None installed by default; opt in with `zot ext install` or `zot --ext`. See [docs/extensions.md](docs/extensions.md).
13- reusable instructions via `SKILL.md` files; see [docs/skills.md](docs/skills.md).
14- no community atm.
15
16## Install
17
18### One-liner (macOS, Linux)
19
20```bash
21curl -fsSL https://www.zot.sh/install.sh | bash
22```
23
24Detects your OS and architecture, downloads the latest release from GitHub, verifies the SHA-256 against the release's `checksums.txt`, extracts the binary, and drops it in `/usr/local/bin`, `~/.local/bin`, or `~/bin`, whichever is writable first. Pass a version or prefix to pin:
25
26```bash
27curl -fsSL https://www.zot.sh/install.sh | bash -s -- v0.0.1 ~/bin
28```
29
30### One-liner (Windows, PowerShell)
31
32```powershell
33iwr -useb https://www.zot.sh/install.ps1 | iex
34```
35
36Drops `zot.exe` into `$HOME\bin` and adds it to the user PATH if missing. Open a fresh terminal afterwards.
37
38### go install
39
40```bash
41go install github.com/patriceckhart/zot/cmd/zot@latest
42```
43
44### From source
45
46```bash
47git clone https://github.com/patriceckhart/zot
48cd zot
49make build        # produces ./bin/zot
50make install      # into $GOPATH/bin
51```
52
53### Prebuilt binaries
54
55Every release on the [releases page](https://github.com/patriceckhart/zot/releases) ships archives for Linux, macOS, and Windows on amd64 and arm64 (except windows/arm64), plus a `checksums.txt` file. Download, verify, `chmod +x`, and drop on your `$PATH`.
56
57## Authenticate
58
59The easiest way is to just run `zot` and type `/login`. The TUI opens even without credentials and walks you through a browser-based login flow.
60
61### Credential lookup order
62
631. `--api-key` flag
642. `ANTHROPIC_API_KEY`, `OPENAI_API_KEY`, `KIMI_API_KEY`, `MOONSHOT_API_KEY`, `DEEPSEEK_API_KEY`, `GEMINI_API_KEY`, or `GOOGLE_API_KEY` env var
653. `$ZOT_HOME/auth.json` (API key or OAuth token; mode 0600)
66
67`$ZOT_HOME` defaults to:
68- macOS: `~/Library/Application Support/zot`
69- Linux: `$XDG_STATE_HOME/zot` or `~/.local/state/zot`
70- Windows: `%LOCALAPPDATA%\zot`
71
72### `/login` flow
73
74Run `zot` and type `/login`. Pick one of two methods:
75
76- **API key**: a small local web server starts on `127.0.0.1:<free-port>`, your browser opens a form, you paste your `sk-ant-...`, `sk-...`, Kimi/Moonshot key, DeepSeek (`sk-...`) key, or Google AI Studio (`AIza...`) Gemini key. zot probes the provider once and saves it to `auth.json` if accepted.
77- **Subscription**: use your Claude Pro/Max, ChatGPT Plus/Pro, or Kimi Code subscription. DeepSeek and Google Gemini do **not** have a subscription login path — the `/login subscription` step only lists Anthropic, OpenAI, and Kimi. For DeepSeek and Google use the API-key flow.
78  - Anthropic and OpenAI pin the browser callback to fixed provider-specific ports (`localhost:53692` for Anthropic, `localhost:1455` for OpenAI) because those are the only ports their auth servers will redirect to.
79  - Anthropic uses the Claude Code OAuth flow. Messages go to `api.anthropic.com` with a bearer token and the Claude Code identity headers.
80  - OpenAI uses the Codex CLI OAuth flow. Messages go to `chatgpt.com/backend-api/codex/responses` with the `chatgpt-account-id` extracted from the returned id_token.
81  - Kimi uses the Kimi Code device-code OAuth flow. zot opens the verification URL, polls until you approve it in the browser, then sends messages to `api.kimi.com/coding/v1` with the Kimi Code identity headers.
82
83> **Note on subscription login.** The OAuth client IDs used are the ones published in Anthropic's Claude Code CLI, OpenAI's Codex CLI, and Kimi Code CLI. Reusing them from a third-party tool may be against their terms of service and may be revoked at any time. Use it at your own risk; the API-key flow is the safe default.
84
85### Token refresh
86
87OAuth access tokens are short-lived (Anthropic ~8h, OpenAI ~30d). zot refreshes them automatically:
88
89- At every credential lookup, zot checks the stored `expiry` and, if past it (with a 60s safety margin), hits the provider's `oauth/token` endpoint with the stored `refresh_token`, persists the new `access_token`, `refresh_token`, and `expiry` back to `auth.json`, and hands the fresh token to the client.
90- The telegram bridge additionally refreshes once per turn so a bot that runs for days keeps working without manual intervention.
91- If the refresh itself fails (the `refresh_token` was revoked, or the account was logged out everywhere), the error bubbles up to the caller: the TUI shows it in the status line, the bot replies with it in your DM. Run `/login` to get a fresh token pair.
92
93All data lives under `$ZOT_HOME`:
94
95```
96$ZOT_HOME/
97├── config.json         # last-used provider/model/theme, saved automatically
98├── auth.json           # api keys and oauth tokens (mode 0600)
99├── sessions/           # jsonl transcripts, one dir per cwd
100├── models-cache.json   # live /v1/models discovery cache (6h ttl)
101├── SYSTEM.md           # optional: replaces the default system prompt
102├── skills/             # optional: user SKILL.md files (opt in with --with-skills)
103├── extensions/         # installed extensions, one dir per extension
104└── logs/               # app log files
105```
106
107Drop a `SYSTEM.md` in `$ZOT_HOME` to replace the built-in identity and guidelines for every run. `--system-prompt` still wins per-invocation. Delete the file to revert to the default.
108
109## Changelog on update
110
111The first time you launch a newer zot binary, the TUI shows the GitHub release notes once in a dismissible overlay. Press any key to close. The version is recorded in `config.json`'s `last_changelog_shown` so the same release notes never reappear. Fresh installs don't see a changelog (no upgrade has happened yet). The fetch is best-effort: a network failure or a missing release page silently skips, with another attempt on the next launch.
112
113## Usage
114
115```bash
116zot                              # interactive tui
117zot "fix the failing test"       # tui, pre-filled prompt
118zot -p "list all go files"       # print final text, exit
119zot --json "refactor main.go"    # newline-delimited json events, exit
120zot --continue                   # resume the most recent session for this cwd
121zot --resume                     # pick a session to resume
122zot --list-models                # show supported models
123zot --help
124```
125
126## Flags
127
128| Flag | Description |
129|---|---|
130| `--provider anthropic\|openai\|kimi\|deepseek\|google\|ollama` | Pick the provider. |
131| `--model <id>` | Pick the model (see `--list-models`). |
132| `--api-key <key>` | Override the API key. |
133| `--base-url <url>` | Override the provider base URL (tests, self-hosted). |
134| `--system-prompt <text>` | Replace the default system prompt for this run (also overrides `$ZOT_HOME/SYSTEM.md`). |
135| `--append-system-prompt <text>` | Append text to the system prompt (repeatable). |
136| `--reasoning low\|medium\|high` | Enable reasoning on supported models. |
137| `-c`, `--continue` | Resume the latest session for this cwd. |
138| `-r`, `--resume` | Pick a session to resume. |
139| `--session <path>` | Resume a specific session file. |
140| `--no-session` | Don't read or write session files. |
141| `--cwd <path>` | Use `<path>` as the working directory. |
142| `--no-tools` | Disable all tools. |
143| `--tools <csv>` | Only enable the listed tools. |
144| `--max-steps <n>` | Cap agent loop iterations (default 50). |
145| `-e`, `--ext <path>` | Load an extension from `<path>` for this run (repeatable; wins against installed extensions of the same name). |
146| `--no-ext` | Skip extension discovery for this run. `--ext` still works on top, so `--no-ext --ext ./x` runs only `x`. |
147| `--with-skills` | Also load user-installed skills. Without this, only the built-in skills shipped in the binary are loaded. |
148| `--no-skill` | Disable all skills, including built-ins. No `skill` tool is registered and the system prompt has no skill manifest. |
149| `--no-yolo` | Confirm every tool call before it runs (interactive TUI only). A dialog shows the tool name and a one-line preview of its args with four choices: yes, yes-always-this-tool-this-session, yes-always-this-session, no. Ignored with a stderr warning in print / json / rpc modes, where tools still run freely so scripts and automation keep working. |
150
151## Tools
152
153- `read`: read text files, or inline images (PNG, JPEG, GIF, WebP).
154- `write`: create or overwrite files, making parent directories as needed.
155- `edit`: one or more exact-match replacements in an existing file.
156- `bash`: run a shell command in the session cwd, with merged stdout/stderr and a timeout.
157
158When the sandbox is on (see `/jail`), all four tools refuse paths outside the session cwd.
159
160## Modes
161
162- **Interactive** (default): chat TUI with streaming output, spinner, cost meter, slash commands.
163- **Print**: `zot -p "prompt"` runs the agent to completion and writes only the final assistant text to stdout.
164- **JSON**: `zot --json "prompt"` emits one JSON object per agent event to stdout, newline-delimited. The schema is documented in [docs/rpc.md](docs/rpc.md).
165- **RPC**: `zot rpc` runs as a long-lived child process; commands in on stdin, events and responses out on stdout, both as NDJSON. Designed for embedding zot in third-party apps written in any language. See [docs/rpc.md](docs/rpc.md) for the wire schema and `examples/rpc/{python,node,shell,go}` for working clients.
166
167## Embedding
168
169Two ways to drive zot from another program:
170
171- **Go in-process**: import `github.com/patriceckhart/zot/pkg/zotcore`. One `Runtime` per project; `Prompt(ctx, text, images)` returns a channel of `Event`. Small example in `examples/sdk/`.
172- **Any language, out-of-process**: spawn `zot rpc` as a subprocess and exchange newline-delimited JSON over its stdin/stdout. Wire format and event schema in [docs/rpc.md](docs/rpc.md). Reference clients live under `examples/rpc/`.
173
174Both interfaces share the same event schema, so transcripts captured by one can be replayed through the other.
175
176## Slash commands
177
178Type `/` in the TUI to open the autocomplete popup. Available commands:
179
180| Command | Description |
181|---|---|
182| `/help` | Show key bindings and commands. |
183| `/login` | Log in via API key or subscription (opens a dialog). |
184| `/logout [provider]` | Clear credentials for `anthropic`, `openai`, `kimi`, `deepseek`, `google`, or all when omitted. `/logout kimi` also disables fallback to the official Kimi Code CLI token until you log in to Kimi through zot again. |
185| `/model` | Pick a model from a list (or `/model <id>` to set directly). |
186| `/sessions` | Resume a previous session for this directory. |
187| `/session` | Four ops on the current session: `export` to a portable `.zotsession` file, `import` one back in, `fork` from a past user message into a new branch, `tree` to switch between branches. Opens a picker without an argument; direct forms: `/session export [path]`, `/session import <path>`, `/session fork`, `/session tree`. Default export destination is `~/Downloads`. |
188| `/jump` | Scroll the chat to a previous turn (or `/jump <text>` to filter). |
189| `/btw` | Side chat with full context that doesn't add to the main thread. |
190| `/swarm` | Spawn, monitor, and chat with background subagents. Each runs in parallel with your main session and shares its working directory. |
191| `/skills` | List discovered skills (SKILL.md files) and preview their bodies. |
192| `/compact` | Summarize the transcript into one message to free up context. |
193| `/study` | Run the canned prompt "Read and understand everything in the current directory." so the agent has full project context before you start asking targeted questions. Pass a path — typed, drag-dropped, or selected via `@` — to target a specific file or directory instead: `/study [dir:internal/]`, `/study cmd/zot/main.go`. |
194| `/jail` | Confine tools to the current directory. |
195| `/unjail` | Allow tools to touch paths outside again. |
196| `/reload-ext` | Hot-reload all extensions (re-read manifests, respawn subprocesses, rebuild tool registry). |
197| `/telegram` | Connect, disconnect, or show status of the Telegram bridge (takes `connect` / `disconnect` / `status` as an optional argument; opens a picker without one). When connected, DMs from the paired user become prompts in the running session and the assistant's replies are mirrored back to Telegram. Alias: `/tg`. |
198| `/clear` | Clear the chat transcript. |
199| `/exit` | Exit zot. |
200
201Extension-registered commands appear under a divider at the bottom of the popup, sorted by name.
202
203### `/sessions`
204
205Shows previous sessions for the current working directory, newest first, with timestamp, model, message count, cost, and the first user prompt. Pick one with `up`/`down`, `enter` to resume, `esc` to cancel. zot swaps the current session file for the selected one and replays the full transcript (including tool calls) into the agent. Sessions remember the model they ended on, so resuming picks up on that exact model even if your global default changed.
206
207### `/session`
208
209Four ops on the current session. `/session` alone opens a picker; each is also runnable directly.
210
211- **`/session export [path]`**. Writes the running transcript to a portable `.zotsession` file. Default destination is `~/Downloads/<timestamp>-<session-id>-<prompt-slug>.zotsession`. Pass a path to override; a directory is fine (a dated name is built inside), a bare name gets `.zotsession` appended. The meta's cwd is stripped on the way out so the recipient doesn't see your filesystem layout.
212
213  **What's included.** Only the main chat thread of the running session — messages, tool calls, tool results, compactions, and usage. **`/swarm` subagents are NOT included.** Their transcripts, unix-socket inboxes, and per-agent session files are all machine-local; a `.zotsession` is just a chat transcript and has no way to revive a unix socket on another box. If you want the conversation, copy it out of the dashboard manually.
214- **`/session import <path>`**. Copies a `.zotsession` file into `$ZOT_HOME/sessions/<cwd-hash>/` with a fresh id and the current cwd, then switches the running agent onto it. Imported sessions are first-class: they show up in `/sessions`, `/jump`, and the tree. Drag-drop paths in the editor are accepted (zot strips the surrounding quotes automatically).
215- **`/session fork`**. Opens a turn picker (same shape as `/jump`). Pick any past user message; zot copies every message up to and including that turn into a new session, records `parent` + `fork_point` in the new meta, and switches onto the branch. The parent session stays on disk. Use it to try a different question without polluting the original transcript, or to rewind after the agent went down the wrong path.
216- **`/session tree`**. Shows every session in the current cwd arranged by parent/child relationships, depth-first with indent per level. The current session is tagged `[current]`. Pick any entry to switch into it. Parentless sessions are roots; branches created via `/session fork` nest under whichever session they were forked from. Orphaned children (whose parent file was deleted) still show as roots so they stay discoverable.
217
218### `/jump`
219
220Opens a turn picker for the current session, one row per user prompt, each showing the turn number, how many tools that turn invoked, and the first line of the prompt. `up`/`down` to pick, `enter` to jump, `esc` to cancel. Any printable rune while the picker is open extends a filter; backspace narrows it back. `/jump <text>` pre-applies the filter; if exactly one turn matches, zot jumps straight there without showing the picker.
221
222Jumping is non-destructive. The transcript is untouched, the viewport just scrolls so the chosen turn is at the top. A muted line at the top of the chat reads `viewing turn N of M, pgdn to catch up`. Scroll back to the bottom with `pgdn` (or keep scrolling with the arrow keys) and the indicator goes away.
223
224### `/btw`
225
226Opens a side-chat overlay with the full main session as frozen context, so you can ask quick clarifying questions ("does asyncio.gather() catch exceptions?", "btw the bundle budget is 10MB", "what's the default fetch timeout?") without bloating the main thread.
227
228Each question fires a one-off model call against `system + main transcript + side-chat history so far`. Responses render in the overlay and stay there. When you press `esc` to close, **nothing** has been added to the main session and subsequent main-thread turns don't re-read any of the side-chat exchanges, keeping the running context window lean.
229
230```
231/btw                              # open the overlay, type questions interactively
232/btw does PUT replace the whole resource?
233```
234
235Inside the overlay: `enter` sends, `esc` cancels an in-flight call (or closes the overlay if idle), `ctrl+c` closes immediately. Side-chat exchanges never touch the transcript and aren't persisted to the session file.
236
237### `/swarm`
238
239Background subagents that run alongside your main session. Each one is a separate `zot` subprocess with its own model loop, its own persistent session file, and its own chat in the dashboard — but they all run in **the same working directory as the host**, so they see and edit the same files you do. Spawn one for a side task (“draft the migration”, “investigate this stack trace”, “write the test harness for module X”), keep going in the main thread, check in on it whenever you want.
240
241> **Agents edit the same files you do.** They use the same `read` / `write` / `edit` / `bash` tools as the main agent against the host's working directory. There's no per-agent worktree or branch. If you need parallel edits on isolated checkouts, set that up yourself with `git worktree` outside zot.
242
243```
244/swarm                            # open the dashboard
245/swarm new <task>                 # spawn an agent
246/swarm new --model gpt-5 <task>    # pin the new agent to a specific model
247/swarm logs <id>                  # jump straight into one agent's transcript
248/swarm send <id> <text>           # send a follow-up without opening the dashboard
249/swarm resume                     # pick a stopped agent to bring back
250/swarm resume <id>                # bring a specific agent back
251/swarm kill <id>                  # stop a running agent (its state stays)
252/swarm remove <id>                # delete the agent's session and state
253/swarm list                       # alias for opening the dashboard
254```
255
256**Dashboard (`/swarm` with no arg)** — a list of every agent for the current session, with status, age, and current activity. Keys:
257
258| Key | Action |
259|---|---|
260| `↑` / `↓` | Move cursor between rows. |
261| `enter` | Open the highlighted agent's transcript view. |
262| `n` | Spawn a new agent (opens an inline task editor; inherits the host's current model). |
263| `p` | One-off prompt editor for the selected row (without entering the transcript). |
264| `R` | Resume a stopped agent in place. |
265| `k` | Kill the selected running agent. Its session and state stay so you can resume it later. |
266| `r` | Remove the selected agent entirely (session + meta gone). |
267| `esc` | Close the dashboard. |
268
269**Inside an agent's transcript** — a chat overlay with an always-on inline composer at the bottom. The conversation flows above it; type and `enter` to send a follow-up. The view auto-follows streaming output and shows an inline spinner with the agent's current activity (`thinking`, `tool: edit_file`, etc.) while it's busy. `esc` returns to the dashboard.
270
271**Switching the spawn model from inside the editor** — while composing a task in the `n`-prompt, type `/model` on its own line and `enter`. The standard `/model` picker pops up; pick a model, the picker closes, and the editor reopens with your typed task intact and the new model pinned for the spawn.
272
273**Session scoping** — each agent is stamped with the host session that spawned it and only shows up in that session's dashboard. Swap sessions with `/sessions` and the dashboard re-narrows accordingly. Agents from other sessions keep running in the background and reappear when you switch back.
274
275**Persistence across zot restarts** — every spawn writes a `meta.json` next to its event log and session file under `$ZOT_HOME/swarm/agents/<id>/`. On the next `zot` launch they show up in the dashboard as **detached**; press `R` (or `/swarm resume <id>`) to bring one back. Resumed agents reattach to the same session and inbox socket, so the conversation continues from where it left off.
276
277**Where state lives** — everything per-agent (session file, events log, inbox socket, meta) lives under `$ZOT_HOME/swarm/agents/<id>/`. The agent's actual code edits land directly in your repo; track them with normal `git status` / `git diff`.
278
279**`/session export` does NOT bundle subagents.** A `.zotsession` is just the main chat transcript; per-agent state (session file, unix-socket inbox) is machine-local and doesn't round-trip through a JSONL file. To share what an agent said, copy it out of the transcript view manually.
280
281### `/skills`
282
283Opens a picker listing every discovered SKILL.md file, built-ins hidden. Each row shows the skill name, source, and description. `enter` opens the body inline (scrollable with `up`/`down`/`pgup`/`pgdn`); `esc` goes back. Re-runs discovery each time it opens, so edits to a SKILL.md during a session are reflected immediately.
284
285### `/compact`
286
287Sends the current transcript through the model with a structured summarization prompt. The returned summary replaces the transcript as one synthetic user message, with the last few exchanges kept verbatim for continuity. The status bar's context meter resets. Use it when the context meter creeps past ~80%.
288
289zot also auto-compacts in the background: after any turn that leaves context usage at or above **85%** of the model's window, the agent kicks off a condense pass on its own. You'll see `condensing history, esc to cancel` above the status bar and an `(auto)` tag next to the context percentage; `esc` aborts it without touching the transcript.
290
291### `/jail`
292
293Enforces a sandbox rooted at the cwd shown in the status bar. `read`, `write`, and `edit` resolve their target path (including through symlinks) and refuse anything outside the sandbox. `bash` refuses obvious escape patterns: `sudo`, `rm -rf /`, leading `cd /`, `cd ..`, `cd ~`, `chmod -R`, `dd of=/`, and similar. The status bar shows `jailed, ~/your/cwd` while active.
294
295This is a guardrail against accidents, not a hard security boundary. If you need real isolation, run zot under docker or a proper sandbox.
296
297## Sessions
298
299Every interactive or print/json run (unless `--no-session`) writes a JSONL transcript under `$ZOT_HOME/sessions/<cwd-hash>/`. Resume any of them with `--continue`, `--resume`, `--session <path>`, or interactively via `/sessions` inside the TUI. Empty sessions (the user exited without prompting) are deleted on close so the list stays tidy.
300
301## Models
302
303`--list-models` or the `/model` picker shows the full catalog. Three sources:
304
305- **Catalog**: models baked into zot, always available.
306- **Live**: IDs discovered from `GET /v1/models` using your stored API key (cached for 6h in `$ZOT_HOME/models-cache.json`, refreshed in the background on startup).
307- **Speculative**: IDs that appear in the upstream generator but aren't live on the public API yet. They'll 404 today and start working the moment the provider ships them.
308
309The context meter in the status line uses the model's advertised context window to show how much of it your last turn consumed.
310
311### Model fallback (rescue)
312
313When a turn fails because of a recoverable provider error — expired token (`401`), permission denied (`403`), rate limit (`429`), provider outage (`502`/`503`/`504`), or a transient network failure — zot opens a **rescue** picker over the chat instead of just painting a red banner.
314
315The picker is the same vertical list / fuzzy filter UI as `/model`, but it only shows models from providers you're currently logged in to (env vars, `auth.json`, Kimi CLI fallback, ollama). The failed model is excluded. Press `↑`/`↓` to choose, `enter` to retry the **same prompt** on the new model, `esc` to dismiss.
316
317Before the actual provider request fires, the OpenAI / Anthropic / Kimi / DeepSeek / Google / OpenAI-Codex clients also do up to two silent retries with short backoff (250ms, 750ms) on `502`/`503`/`504` and connection-reset / EOF-before-headers errors. Most edge-proxy blips disappear without you ever seeing the rescue picker.
318
319A rescue retry always **drops launch-time `--api-key` and `--base-url`** before rebuilding the agent. Those overrides are usually the reason the rescue triggered (bad key, typo'd base URL, corporate gateway only valid for the originally-picked provider), so the retry re-resolves credentials from env vars / `auth.json` / provider defaults instead. Use `/model` if you want overrides to stick.
320
321No configuration is required — the candidate list is built dynamically from your active credentials. Bad-request / context-length / serialization errors are NOT routed to the rescue picker, because switching models won't fix them; those still surface as a normal error.
322
323### Custom models
324
325Place a `models.json` in `$ZOT_HOME` (macOS: `~/Library/Application Support/zot/`, Linux: `~/.local/state/zot/`) to add models that aren't in the baked-in catalog or to override existing entries:
326
327```json
328{
329  "providers": {
330    "openai": {
331      "models": [
332        {
333          "id": "gpt-5.5",
334          "name": "GPT-5.5",
335          "reasoning": true,
336          "contextWindow": 400000,
337          "maxTokens": 128000
338        }
339      ]
340    }
341  }
342}
343```
344
345Supported fields per model: `id` (required), `name`, `reasoning`, `contextWindow`, `maxTokens`, `baseUrl`, `priceInput`, `priceOutput`, `priceCacheRead`, `priceCacheWrite`.
346
347Provider keys are normalized: `openai-codex` and `openai-responses` map to `openai`, `anthropic-messages` maps to `anthropic`, `moonshot`, `moonshot-ai`, and `kimi-code` map to `kimi`, and `deepseek-chat` and `deepseek-ai` map to `deepseek`.
348
349User-defined models show `source: user` in `--list-models` and take precedence over both the baked-in catalog and live-discovered models. Missing or invalid files are silently ignored.
350
351### Kimi Code
352
353zot has built-in Kimi support through Kimi's OpenAI-compatible chat API.
354
355```bash
356zot --provider kimi
357```
358
359By default this uses:
360
361- model: `kimi-for-coding`
362- base URL: `https://api.kimi.com/coding/v1`
363
364Credential lookup order for Kimi:
365
3661. `--api-key`
3672. `KIMI_API_KEY`
3683. `MOONSHOT_API_KEY`
3694. `$ZOT_HOME/auth.json`
3705. the official Kimi Code CLI token at `~/.kimi/credentials/kimi-code.json`, unless disabled by `/logout kimi`
371
372Use `/login` for either API-key login or Kimi Code subscription login. The subscription flow uses Kimi Code's device-code OAuth flow: zot opens the verification URL, waits for browser approval, stores the token in `auth.json`, and refreshes it automatically.
373
374For direct Moonshot API keys or a custom compatible endpoint:
375
376```bash
377zot --provider kimi --model kimi-k2-0905-preview --base-url https://api.moonshot.ai/v1 --api-key "$KIMI_API_KEY"
378```
379
380You can add additional Kimi/Moonshot model IDs to `models.json` under the `kimi` provider.
381
382### DeepSeek
383
384zot has built-in DeepSeek support through DeepSeek's OpenAI-compatible chat API.
385
386```bash
387zot --provider deepseek
388```
389
390By default this uses:
391
392- model: `deepseek-v4-pro`
393- base URL: `https://api.deepseek.com/v1`
394
395Catalog ships with `deepseek-v4-pro` (reasoning) and `deepseek-v4-flash`. These are exactly the IDs returned by `GET https://api.deepseek.com/models` today. You can add additional model IDs to `models.json` under the `deepseek` provider.
396
397Credential lookup order for DeepSeek:
398
3991. `--api-key`
4002. `DEEPSEEK_API_KEY`
4013. `$ZOT_HOME/auth.json`
402
403Use `/login` and pick **api key** to paste a DeepSeek key. zot probes `/v1/models` once and stores the key under `deepseek` in `auth.json`.
404
405> **Auth model: API key only.** DeepSeek does not offer a subscription OAuth flow. The `/login subscription` step lists only Anthropic, OpenAI, and Kimi; DeepSeek shows up only under `/login → api key`.
406
407> **Text only at the wire level.** DeepSeek's chat-completions endpoint currently rejects the multimodal content schema (`unknown variant image_url, expected text`). When the active provider is `deepseek`, zot silently drops `ImageBlock` parts from outgoing user/tool messages and keeps only the text. Switching back to a vision-capable model (Claude, GPT-4o/5, Gemini) re-sends the image normally because the session file still stores it.
408
409For a custom-compatible endpoint (mirror, gateway, self-host):
410
411```bash
412zot --provider deepseek --base-url https://my-deepseek-mirror.example.com/v1 --api-key "$DEEPSEEK_API_KEY"
413```
414
415### Google Gemini
416
417zot has built-in Google Gemini support through the [AI Studio Generative Language API](https://aistudio.google.com/).
418
419```bash
420zot --provider google
421```
422
423By default this uses:
424
425- model: `gemini-2.5-pro`
426- base URL: `https://generativelanguage.googleapis.com`
427
428Catalog ships with `gemini-2.5-pro`, `gemini-2.5-flash`, `gemini-2.5-flash-lite`, `gemini-2.0-flash`, and `gemini-2.0-flash-lite`. Live discovery against `/v1beta/models` adds anything else your key can see.
429
430Credential lookup order for Google:
431
4321. `--api-key`
4332. `GEMINI_API_KEY`
4343. `GOOGLE_API_KEY`
4354. `$ZOT_HOME/auth.json`
436
437Use `/login` and pick **api key** to paste an AI Studio key. zot probes `/v1beta/models` once and stores the key under `google` in `auth.json`.
438
439> **Auth model: API key only.** Google does not issue OAuth tokens for consumer Gemini Advanced / Google One AI Premium subscriptions, so there is no "log in with your Google subscription" flow. Programmatic access requires either an AI Studio API key (this provider) or a Vertex AI / GCP service-account credential (not yet wired up in zot). The `/login subscription` step quietly downgrades to the api-key form when you pick Google so you don't end up in a dead end.
440
441> **Free-tier rate limits.** AI Studio's free tier has tight per-minute and per-day caps that vary by model: `gemini-2.5-pro` is the strictest (a few requests per minute, ~50 per day), Flash and Flash-Lite are far more generous. If a Pro turn 429s with `"You exceeded your current quota"` while Flash on the same key still works, you've hit the Pro free-tier RPD. Either switch to Flash for agent loops, or [enable billing](https://aistudio.google.com/app/apikey) on your AI Studio project to flip the same key from free to pay-as-you-go pricing (`$1.25/M` input, `$10/M` output for Pro).
442
443Reasoning levels (`--reasoning low|medium|high`) map differently per generation: 2.5 family uses `thinkingBudget` token budgets per model (Pro caps at 32k, Flash at 24k); Gemini 3.x uses the `thinkingLevel` enum (`MINIMAL`/`LOW`/`MEDIUM`/`HIGH`), with Gemini-3-Pro pinned to `LOW` minimum and `HIGH` for any "medium" or "high" request. 2.0-family models have no thinking config at all.
444
445You can add additional Gemini model IDs to `models.json` under the `google` provider.
446
447### Local models with ollama
448
449zot works with [ollama](https://ollama.com) out of the box. Ollama serves an OpenAI-compatible API locally, so any model you have pulled works with zot.
450
451Quick start:
452
453```bash
454ollama pull qwen3.5:4b
455zot --provider ollama --model qwen3.5:4b
456```
457
458That's it. No API key needed for local models. zot defaults to `http://localhost:11434`.
459
460For a remote ollama instance or one behind auth:
461
462```bash
463zot --provider ollama --model llama3 --base-url https://my-server.com/v1 --api-key my-token
464```
465
466You can also add models to your `models.json` so you don't need flags every time:
467
468```json
469{
470  "providers": {
471    "ollama": {
472      "models": [
473        {
474          "id": "qwen3.5:4b",
475          "name": "Qwen 3.5 4B",
476          "contextWindow": 32768,
477          "maxTokens": 8192
478        }
479      ]
480    }
481  }
482}
483```
484
485The `ollama` provider uses the OpenAI chat completions protocol internally, so it also works with any OpenAI-compatible server (vLLM, LM Studio, LocalAI, etc.).
486
487## Inline images
488
489When a tool returns an image (for example `read` on a PNG), zot renders it inline on terminals that support it: **Ghostty**, **Kitty**, **iTerm2**, **WezTerm**. On other terminals you see a text placeholder with MIME type, pixel dimensions, and byte size. Control with the `ZOT_INLINE_IMAGES` env var:
490
491| Value | Effect |
492|---|---|
493| unset (default) | Auto-detect based on `TERM_PROGRAM`. |
494| `iterm`, `iterm2` | Force the iTerm2 OSC 1337 protocol. |
495| `kitty` | Force the Kitty graphics protocol. |
496| `off`, `none` | Always use the text placeholder. |
497
498Frames containing images are full-repainted (no differential diff) to prevent stale image pixels from lingering through scroll. That costs one terminal flash per image-containing frame; set `ZOT_INLINE_IMAGES=off` if that bothers you.
499
500## Queued messages
501
502You can keep typing while the agent is working. Pressing `enter` during a turn queues the message instead of interrupting: it shows up above the status bar as `sliding in: <text>` and is delivered as the next user turn the moment the current one finishes. Queue as many as you want; they run in order. `esc` cancels the active turn and drops the queue so a runaway turn doesn't flood you with stale follow-ups; `ctrl+c` while busy arms the exit hint instead of interrupting, a second `ctrl+c` within two seconds exits zot.
503
504To recover the most recently queued message back into the editor (to tweak it before it runs), press `Option+↑`. In VS Code's integrated terminal that chord doesn't survive xterm.js's macOS key handling — use `Option+Shift+↑` there. zot's hint line under the sliding-in queue adapts automatically based on `$TERM_PROGRAM`.
505
506Slash commands also work while the agent is busy. Read-only ones (`/help`, `/jump`, `/btw`, `/sessions`, `/skills`, `/jail`, `/unjail`, `/exit`) take effect immediately. Destructive ones (`/clear`, `/compact`, `/login`, `/logout`, `/model`, `/reload-ext`) cancel the active turn first and then run.
507
508
509## Keys (interactive mode)
510
511### Input
512
513| Key | Action |
514|---|---|
515| `enter` | Submit (queued if the agent is busy). |
516| `alt+enter` | Newline. |
517| `tab` | Complete the selected slash command. |
518| `esc` | Cancel the current turn (while busy); clear input (while idle). |
519| `ctrl+c` | Clear the input and queue (while idle) or arm the exit hint (while busy). Press again within 2s to exit. Use `esc` to cancel a running turn. |
520| `ctrl+d` | Exit on empty input. |
521| `ctrl+l` | Redraw the screen. |
522| `ctrl+o` | Expand or collapse long tool results (read, write, edit, bash outputs over ~12 lines). |
523| `@` | Open the file picker. Browse files and directories in the working directory. |
524
525### File picker (`@`)
526
527| Key | Action |
528|---|---|
529| `@` | Open the file picker (type after a space or at the start of input). |
530| `up`, `down` | Navigate the file list. |
531| `right` | Open the selected directory. |
532| `left` | Go back to the parent directory. |
533| `enter` | Select the file or directory and insert it as a chip (`[file:name]` or `[dir:name/]`). |
534| `esc` | Close the file picker. |
535
536Type `@` followed by a filter string to narrow the list (e.g. `@read` shows only entries containing "read"). Selected files are inserted as compact chips that expand to the full path on submit. Dragged-and-dropped files and directories also collapse to chips automatically.
537
538### Editor line navigation
539
540| Key | Action |
541|---|---|
542| `ctrl+a`, `ctrl+e` | Jump to start or end of line. |
543| `alt+left`, `alt+right` | Jump one word back or forward. |
544| `ctrl+u`, `ctrl+k` | Delete to start or end of line. |
545| `ctrl+w`, `alt+backspace` | Delete the previous word. |
546| `up`, `down` (editor non-empty) | Cycle through prompt history. |
547
548### Chat scroll
549
550| Key | Action |
551|---|---|
552| `pgup`, `pgdn` | Scroll one page up or down. |
553| `up`, `down` (editor empty) | Scroll three lines up or down. This is how the mouse wheel reaches the scroll logic on most terminals. |
554
555## Extensions
556
557zot can be extended in any language via a subprocess + JSON-RPC protocol. Extensions can register slash commands, expose tools to the model, intercept tool calls (block or rewrite args), gate whole turns before the model is called, and rewrite the assistant's visible text before it reaches the user. None are installed by default; opt in explicitly. Hot-reload any time with `/reload-ext`.
558
559### Install and manage
560
561```bash
562zot ext install <path|git-url>   # copy / clone into $ZOT_HOME/extensions/
563zot ext list                      # show installed extensions
564zot ext logs <name> [-f]          # cat or tail the extension's stderr log
565zot ext enable <name>             # re-enable a disabled extension
566zot ext disable <name>            # disable without removing
567zot ext remove <name>             # delete an extension directory
568```
569
570For development, point `zot --ext <path>` at a working directory and skip the install step entirely. Repeatable; takes precedence over installed extensions of the same name.
571
572### Reference
573
574`examples/extensions/` ships reference implementations in Go, TypeScript, Node, and shell. See [docs/extensions.md](docs/extensions.md) for the full protocol, the SDK API (`pkg/zotext`), and the phase roadmap.
575
576## Skills
577
578A skill is a per-folder `SKILL.md` file with a YAML frontmatter header. zot discovers skills at startup, surfaces their names in the system prompt, and exposes a built-in `skill` tool the model uses to load the body on demand.
579
580By default only the built-in skills shipped with the zot binary are loaded. Pass `--with-skills` to also load user-installed skills from:
581
582- `./.zot/skills/<name>/SKILL.md` (project)
583- `$ZOT_HOME/skills/<name>/SKILL.md` (global)
584- `./.claude/skills/<name>/SKILL.md`, `~/.claude/skills/<name>/SKILL.md` (Claude-compatible layout)
585- `./.agents/skills/<name>/SKILL.md`, `~/.agents/skills/<name>/SKILL.md` (agent-compatible layout)
586
587See [docs/skills.md](docs/skills.md) for the frontmatter fields, authoring tips, and example skills under `examples/skills/`.
588
589## Telegram bot (bridge)
590
591zot can run as a telegram bot so you can DM it from your phone. Two ways to run it: **from inside the TUI** (the running session mirrors into Telegram) or **as a standalone background daemon** (a headless bot with its own independent agent).
592
593### From inside the TUI
594
595Type `/telegram` in the running TUI to open a picker with **connect**, **disconnect**, and **status**. When connected:
596
597- DMs from the paired user become prompts in the **same** session you're typing in, so you can continue a conversation from the terminal on your phone and back again.
598- Messages you type in the TUI are mirrored into the Telegram thread prefixed `you: …` and the assistant's replies come back prefixed `zot: …`, so the Telegram chat stays a complete record of both sides of the conversation.
599- Messages sent from Telegram show up as your own bubble in Telegram (no mirror) and the assistant's reply to them comes back bare (no prefix).
600- The status bar shows a `- tg -` tag while the bridge is active.
601- `/telegram connect` / `/telegram disconnect` / `/telegram status` (or `/tg`) also work as direct commands without the picker.
602
603The in-TUI bridge refuses to start while the standalone daemon (below) is running, since two concurrent long-poll consumers of the same bot race on every update and silently drop messages.
604
605### Standalone daemon
606
607For headless servers or long-running bots unattached to a TUI:
608
609```bash
610zot telegram-bot setup     # paste a BotFather token, verify, save
611zot telegram-bot run       # foreground: long-poll in this terminal (ctrl+c to stop)
612zot telegram-bot start     # background: detach and return immediately
613zot telegram-bot stop      # SIGTERM the background bot (SIGKILL after 5s)
614zot telegram-bot logs -f   # tail $ZOT_HOME/logs/bot.log (omit -f to just cat)
615zot telegram-bot status    # config (token masked) + running/stopped
616zot telegram-bot reset     # forget the token and paired user
617# short alias: `zot tg ...` is accepted for every subcommand
618```
619
620The background flavor writes the child's PID to `$ZOT_HOME/bot.pid` and redirects stdout and stderr to `$ZOT_HOME/logs/bot.log`. `zot telegram-bot stop` reads that PID, sends SIGTERM, waits up to five seconds, then escalates to SIGKILL if the child is still alive. Running two instances at once is refused at startup.
621
622> **Use the installed binary for `start`.** `go run ./cmd/zot telegram-bot start` won't work. `go run` builds a binary in a temp directory and deletes it when it exits, which kills the detached child. Run `make install` (or `go build`) first and invoke the installed binary.
623
624Setup flow:
625
6261. Talk to [@BotFather](https://t.me/BotFather) on telegram, run `/newbot`, copy the token it gives you.
6272. Run `zot telegram-bot setup` and paste the token when prompted.
6283. Run `zot telegram-bot run` in the directory you want the agent to operate in.
6294. Open your bot on telegram, send `/start`. The first user to do this claims the bridge (stored as `allowed_user_id`); every other user is rejected.
630
631From then on, any DM you send is forwarded to the agent as a user prompt. Attached photos or `image/*` documents are downloaded and passed to vision-capable models. In-bot telegram commands: `/help`, `/status`, `/stop` (cancel the current turn). Config lives in `$ZOT_HOME/bot.json` (mode 0600).
632
633Bot mode respects the usual zot flags: `--provider`, `--model`, `--cwd`, `--reasoning`, `--continue`, `--no-session`, `--no-tools`, and so on. Run `zot tg run -c --model claude-opus-4-1` to resume the latest session on Opus, for example.
634
635## Development
636
637```bash
638make build     # build ./bin/zot
639make test      # go test -race ./...
640make lint      # go vet + gofmt check
641make fmt       # gofmt -w .
642make release   # cross-compile linux/darwin/windows on amd64 and arm64
643```
644
645Source layout:
646
647```
648cmd/zot/                      main()
649internal/agent/               cli wiring, arg parsing, system prompt, config
650internal/agent/extensions/    extension subprocess manager
651internal/agent/modes/         interactive tui, print, json, dialogs
652internal/agent/tools/         read, write, edit, bash, sandbox
653internal/auth/                credential store, api-key probe, oauth, login server
654internal/core/                agent loop, sessions, cost tracking
655internal/extproto/            extension wire-format types
656internal/provider/            anthropic + openai-compatible streaming clients, model catalog
657internal/skills/              skill discovery, frontmatter parser, skill tool
658internal/tui/                 terminal raw-mode, input parser, editor, renderer, markdown, view
659pkg/zotcore/                  public Go SDK for embedding zot in-process
660pkg/zotext/                   public Go SDK for writing extensions
661```
662
663## License
664
665MIT