@just-every/code 0.4.1 → 0.4.3
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.
- package/README.md +43 -17
- package/package.json +6 -6
package/README.md
CHANGED
|
@@ -11,23 +11,41 @@
|
|
|
11
11
|
**Code** is a fast, local coding agent for your terminal. It's a community-driven fork of `openai/codex` focused on real developer ergonomics: Browser integration, multi-agents, theming, and reasoning control — all while staying compatible with upstream.
|
|
12
12
|
|
|
13
13
|
 
|
|
14
|
-
##
|
|
14
|
+
## What's new in v0.4.0 (October 26, 2025)
|
|
15
|
+
|
|
16
|
+
- **Auto Drive runs the whole play** – hand `/auto` a task and it now plans, coordinates agents, reruns checks, and recovers from hiccups without babysitting.
|
|
17
|
+
- **Unified settings overlay** – `/settings` centralizes limits, model routing, themes, and CLI integrations so you can audit configuration in one place.
|
|
18
|
+
- **Card-based activity stream** – Agents, browser sessions, web search, and Auto Drive render as compact cards with drill-down overlays for full logs.
|
|
19
|
+
- **Turbocharged performance** – History rendering and streaming were optimized to stay smooth even during long multi-agent sessions.
|
|
20
|
+
- **Smarter agent routing** – Mix and match orchestrator CLIs (Claude, Gemini, GPT-5, Qwen, and more) per `/plan`, `/code`, or `/solve` run.
|
|
15
21
|
|
|
16
|
-
|
|
17
|
-
- 📝 **Diff Viewer** - Side-by-side diffs with syntax highlighting
|
|
18
|
-
- 🤖 **Multi-Agent Commands** - /plan, /solve, /code with agent panels
|
|
19
|
-
- 🎨 **Theme System** - /themes with live preview and accessibility
|
|
20
|
-
- 🧠 **Reasoning Control** - /reasoning for dynamic effort adjustment
|
|
21
|
-
- 🔌 **MCP support** – Extend with filesystem, DBs, APIs, or your own tools.
|
|
22
|
-
- 🔒 **Safety modes** – Read-only, approvals, and workspace sandboxing.
|
|
23
|
-
- 🔁 **Backwards compatible** – Reads both `~/.code/*` (primary) and legacy `~/.codex/*`; writes only to `~/.code/*`
|
|
22
|
+
Read the full notes in `release-notes/RELEASE_NOTES.md`.
|
|
24
23
|
|
|
25
24
|
 
|
|
26
|
-
|
|
27
|
-
|
|
25
|
+
## Why Code
|
|
26
|
+
|
|
27
|
+
- 🚀 **Auto Drive orchestration** – Multi-agent automation that now self-heals and ships complete tasks.
|
|
28
|
+
- 🌐 **Browser Integration** – CDP support, headless browsing, screenshots captured inline.
|
|
29
|
+
- 🤖 **Multi-agent commands** – `/plan`, `/code` and `/solve` coordinate multiple CLI agents.
|
|
30
|
+
- 🧭 **Unified settings hub** – `/settings` overlay for limits, theming, approvals, and provider wiring.
|
|
31
|
+
- 🎨 **Theme system** – Switch between accessible presets, customize accents, and preview live via `/themes`.
|
|
32
|
+
- 🔌 **MCP support** – Extend with filesystem, DBs, APIs, or your own tools.
|
|
33
|
+
- 🔒 **Safety modes** – Read-only, approvals, and workspace sandboxing.
|
|
34
|
+
|
|
35
|
+
<p align="center">
|
|
36
|
+
<strong>Introducing Auto Drive</strong><br>
|
|
37
|
+
<a href="https://youtu.be/Du3p7Q8_eUo">
|
|
38
|
+
<img src="docs/screenshots/video-auto-drive-play.jpg" alt="Play Introducing Auto Drive video" width="100%">
|
|
39
|
+
</a>
|
|
40
|
+
</p>
|
|
41
|
+
|
|
42
|
+
<p align="center">
|
|
43
|
+
<strong>Multi-Agent Support</strong><br>
|
|
44
|
+
<a href="https://youtu.be/1_X8O8W2X0M">
|
|
45
|
+
<img src="docs/screenshots/video-v03-play.jpg" alt="Play Multi-Agent Support video" width="100%">
|
|
46
|
+
</a>
|
|
47
|
+
</p>
|
|
28
48
|
|
|
29
|
-
| <br><img src="docs/screenshots/browser.png" alt="Browser control" width="100%"><br>Browser control | <br><img src="docs/screenshots/agents.png" alt="Assist with Claude & Gemini" width="100%"><br>Assist with Claude & Gemini |
|
|
30
|
-
|:--:|:--:|
|
|
31
49
|
|
|
32
50
|
|
|
33
51
|
 
|
|
@@ -51,7 +69,6 @@ Note: If another tool already provides a `code` command (e.g. VS Code), our CLI
|
|
|
51
69
|
**Authenticate** (one of the following):
|
|
52
70
|
- **Sign in with ChatGPT** (Plus/Pro/Team; uses models available to your plan)
|
|
53
71
|
- Run `code` and pick "Sign in with ChatGPT"
|
|
54
|
-
- Stores creds locally at `~/.code/auth.json` (still reads legacy `~/.codex/auth.json` if present)
|
|
55
72
|
- **API key** (usage-based)
|
|
56
73
|
- Set `export OPENAI_API_KEY=xyz` and run `code`
|
|
57
74
|
|
|
@@ -109,6 +126,15 @@ qwen --version
|
|
|
109
126
|
/code "Show dark mode when I feel cranky"
|
|
110
127
|
```
|
|
111
128
|
|
|
129
|
+
### Auto Drive
|
|
130
|
+
```bash
|
|
131
|
+
# Hand off a multi-step task; Auto Drive will coordinate agents and approvals
|
|
132
|
+
/auto "Refactor the auth flow and add device login"
|
|
133
|
+
|
|
134
|
+
# Resume or inspect an active Auto Drive run
|
|
135
|
+
/auto status
|
|
136
|
+
```
|
|
137
|
+
|
|
112
138
|
### General
|
|
113
139
|
```bash
|
|
114
140
|
# Try a new theme!
|
|
@@ -189,7 +215,7 @@ Code supports MCP for extended capabilities:
|
|
|
189
215
|
- **API integrations**: Connect to external services
|
|
190
216
|
- **Custom tools**: Build your own extensions
|
|
191
217
|
|
|
192
|
-
Configure MCP in `~/.code/config.toml`
|
|
218
|
+
Configure MCP in `~/.code/config.toml` Define each server under a named table like `[mcp_servers.<name>]` (this maps to the JSON `mcpServers` object used by other clients):
|
|
193
219
|
|
|
194
220
|
```toml
|
|
195
221
|
[mcp_servers.filesystem]
|
|
@@ -230,7 +256,7 @@ model_reasoning_summary = "detailed"
|
|
|
230
256
|
|
|
231
257
|
### Environment variables
|
|
232
258
|
|
|
233
|
-
- `
|
|
259
|
+
- `CODE_HOME`: Override config directory location
|
|
234
260
|
- `OPENAI_API_KEY`: Use API key instead of ChatGPT auth
|
|
235
261
|
- `OPENAI_BASE_URL`: Use alternative API endpoints
|
|
236
262
|
- `OPENAI_WIRE_API`: Force the built-in OpenAI provider to use `chat` or `responses` wiring
|
|
@@ -295,7 +321,7 @@ Using OpenAI, Anthropic or Google services through Code means you agree to **the
|
|
|
295
321
|
- If you configure other model providers, you're responsible for their terms.
|
|
296
322
|
|
|
297
323
|
### Privacy
|
|
298
|
-
- Your auth file lives at `~/.code/auth.json`
|
|
324
|
+
- Your auth file lives at `~/.code/auth.json`
|
|
299
325
|
- Inputs/outputs you send to AI providers are handled under their Terms and Privacy Policy; consult those documents (and any org-level data-sharing settings).
|
|
300
326
|
|
|
301
327
|
### Subject to change
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@just-every/code",
|
|
3
|
-
"version": "0.4.
|
|
3
|
+
"version": "0.4.3",
|
|
4
4
|
"license": "Apache-2.0",
|
|
5
5
|
"description": "Lightweight coding agent that runs in your terminal - fork of OpenAI Codex",
|
|
6
6
|
"bin": {
|
|
@@ -35,10 +35,10 @@
|
|
|
35
35
|
"prettier": "^3.3.3"
|
|
36
36
|
},
|
|
37
37
|
"optionalDependencies": {
|
|
38
|
-
"@just-every/code-darwin-arm64": "0.4.
|
|
39
|
-
"@just-every/code-darwin-x64": "0.4.
|
|
40
|
-
"@just-every/code-linux-x64-musl": "0.4.
|
|
41
|
-
"@just-every/code-linux-arm64-musl": "0.4.
|
|
42
|
-
"@just-every/code-win32-x64": "0.4.
|
|
38
|
+
"@just-every/code-darwin-arm64": "0.4.3",
|
|
39
|
+
"@just-every/code-darwin-x64": "0.4.3",
|
|
40
|
+
"@just-every/code-linux-x64-musl": "0.4.3",
|
|
41
|
+
"@just-every/code-linux-arm64-musl": "0.4.3",
|
|
42
|
+
"@just-every/code-win32-x64": "0.4.3"
|
|
43
43
|
}
|
|
44
44
|
}
|