Plan mode removes the guesswork. When the change is too big for a quick fix: • Your code stays untouched • Plan it first, then review it • Apply exactly as planned No surprises, just clarity.

We ran a test between the new Qwen3.8-Max, Opus 5 and GPT-5.6 Sol. 3 open models. same prompt. one-shot with the /design command. Reviewed gameplay, UX/UI and cost. 🔹 Qwen3.8-Max → 9/10 · $0.0480 🔹 Opus 5 → 9/10 · $0.253 🔹 GPT-5.6 Sol → 9/10 · $0.150

Create custom slash commands and use them frequently. 🔹 Save a prompt as a markdown file 🔹 Define its scope per project or user level 🔹 Run it anytime with dynamic arguments $ /<command-name> [arguments]

GPT-5.6 Terra and Luna are 50% off in Command Code. Get 2X requests for every dollar of credit you spend Luna: - $0.10/M input, $0.60/M output tokens - Available on all plans (Go, Pro, Max, Team) Terra: - $1/M input, $6/M output tokens - Available on Pro, Max, Team For limited time only.

Tested the new DeepSeek V4 Flash, Kimi K3, GLM 5.2. 3 models. 1 prompt. /design command. Reviewed features, UX and cost. 🔹 Kimi K3: 9.5/10 · $0.0740 🔹 GLM 5.2: 9/10 · $0.0480 🔹 DeepSeek V4 Flash: 7/10 · $0.0005 DSV4 is ~150x cheaper, but rough around edges Kimi K3 & GLM 5.2 built a polished game

Inkling-Small by Thinking Machines Lab is now live in Command Code. - $0.50/M input, 1.20/M output tokens - Light-weight model with 12B params - Multimodal: text, image, audio Available on all plans $ cmd --update (to latest) $ cmd --model thinkingmachines/inkling-small

New in v1: Stashing Prompts - `ctrl+s` sets the prompt you are typing aside - press it again and it reappears - no need for ctrl+c and ctrl+v again Never leave a prompt behind, stash it instead.

Stop repeating the same prompt in every session. Agent Skills: an open standard for teaching agents specialized knowledge and workflows. Write it once in a SKILL[.]md file (with references/assets) and reuse it. Save it per project, globally, or in a repo: $ cmd skills add [repo name] --global

Getting lots of requests for our GUI app. Devs are already testing it. Join our Discord.

How's your experience with Qwen 3.7 Flash?

Qwen3.7 Flash is now live in Command Code. 1M context at $0.03/$0.13 per 1M tokens. Available on all plans (Go, Pro, Max, Team).

Bring back any message or code instantly. Use /rewind to pick a checkpoint, then choose what to restore: conversation, code, or both. Never lose your conversation or code again.

We ran a test between Opus 5, GPT-5.6 Sol, and Kimi K3. One prompt. Across all three models. With our /design command. Ranking (DX, features, and cost): 🔹 Opus 5: 10/10 · $0.25 🔹 Kimi K3: 9.5/10 · $0.12 🔹 GPT-5.6 Sol: 7/10 · $0.32

New in Command Code: dont-ask permission mode. Runs exactly what your allowlist permits Everything else gets denied Built for CI jobs, cron, and unattended runs $ cmd --permission-mode dont-ask No more hung jobs. Opposite of --yolo

Claude Opus 5 is now live in Command Code. - Anthropic's latest thoughtful and proactive model - On frontier intelligence, closet to Fable 5 at half price - Available on plans (Pro, Max, Team) Try now: $ cmd --model claude-opus-5

Ling 3.0 Flash is live in Command Code MoE model built for token-efficient agentic inference at production scale. - Free for all subscribers - 124B params (~5.1B active/token) - 256K context, tool use, implicit caching Try it: $ cmd --model inclusionai/ling-3.0-flash-free

We ran a test between Fable 5, GPT-5.6 Sol, and Kimi K3. One prompt. Across all three models. With our /design command. Kimi K3 performed well across all three, We reviewed gameplay, UX, and UI. Ranking (DX and cost): → Kimi K3: 9.5/10 · $0.12 → Fable 5: 9/10 · $0.60 → GPT-5.6 Sol: 7/10 · $0.32

You can pull and push your taste packages. Push what it learned in a project: $ npx taste push Pull it into another project: $ npx taste pull username/project-name If there's an conflict, taste pull can merge, overwrite or skip these changes. So what's your coding taste? Share in comments below.

We ran a test between GPT-5.6 Sol, Kimi K3, and Gemini 3.6 Flash. One prompt across all three models, using our /design command. Ranking (DX, features, and cost): 🔹 Kimi K3: 10/10 · $0.10 🔹 GPT-5.6 Sol: 9/10 · $0.52 🔹 Gemini 3.6 Flash: 6/10 · $0.45

Gemini 3.6 Flash and Gemini 3.5 Flash-Lite are live. Available on Pro, Max and Team plans Gemini 3.6 Flash: 🔹 Better knowledge work & multimodal performance. 17% fewer output Gemini 3.5 Flash-Lite: 🔹 Fast, most cost effective, 350 output tokens/second Try now with /model in the Command Code CLI.

Tested Kimi K3 to build a 3D portfolio website. - Used the /design command. - Built in 3 prompts: build, continue, then recolor - Full session cost: $0.57 The interesting part: Recolor replaced and added an entirely new palette with the same layout and animations in seconds.

We tested Kimi K3 to build liquid metal text that wraps with your cursor in Three[.]js - Used the /design command - 3 prompts to get this effect - Full session cost: $0.77 Open models deliver their best when used with the right harness.

You can create custom agents for yourself or for your team. Run /agents, describe what it should do, which tools it can access, and Command Code handles the rest. Commit it to your repo and your team can use them too. Share what agents have you built with the Command Code CLI.

We compare Kimi K3 with Fable 5. We use our /design command for this test. The result is insanely good for only $0.035. The section and scroll animations, plus the mouse-hover animation in the footer, are outstanding. What you’re cooking with /design using Kimi K3?

Kimi K3 vs GPT-5.6 Sol vs Fable 5 Same /design prompt across all three. Kimi K3 nailed the design and pace. Fable 5 & GPT-5.6 Sol both play too fast (not sped up). Ranking & cost: → Kimi K3: 9.5 · $0.030 → Fable 5: 7.5 · $0.38 → GPT-5.6 Sol: 7 · $0.11

Your agents forget everything between sessions. agents[.]md fixes that. Run /init to generate project memory in Command Code. Scans your project and files to understand: - Your architecture decisions - Your logging patterns - Your team conventions New session, new agent, same context every time.

GPT-5.6 Sol vs Fable 5 vs Grok 4.5 vs Muse Spark 1.1 We tested all four using the Chrome Dino game via the /design command. Parameters reviewed: Dino speed, Behavior, Scoring, Smoothness, Gameplay level Scores (gameplay): - GPT-5.6 Sol: 9/10 - Fable 5: 8/10 - Grok 4.5: 7/10 - Muse Spark 1.1: 6/10

Command Code has permission modes for your coding workflows. 3 modes: Default: Approve each action (diff + reasoning) Accept Edits: Lets Command edit files in full flow Plan Mode: Read-only, research before making any changes Switch them anytime with Shift+Tab. Which mode do you use the most?

GPT-5.6 Sol vs Fable 5 vs Grok 4.5 Built a racing game with a UX focus with /design. Reviewed Speed, crash behavior, scoring, UX. Results Grok 4.5: 9/10, best overall performance Fable 5: 8/10, a close second, matching Grok 4.5 GPT-5.6 Sol: 7/10 best UI, weak gameplay (excess drift & friction)

Fable 5 vs Grok 4.5 vs GPT 5.5 Used Command Code /design, and the exact same prompt. RESULT: Grok 4.5 genuinely plays like a polished mobile game. Fable 5 and GPT 5.5 feel too fast. Everything feels rushed, lacks finish. Ranking based on DX: → Grok 4.5: 9/10 → Fable 5: 7.5/10 → GPT 5.5: 7/10