Neura Market
Advance Minimax M2 Cursor Rules logo

Advance Minimax M2 Cursor Rules

Free

Agentic-first Cursor Rules powered by MiniMax M2 — clarify-first prompting, interleaved thinking, and full tool orchestration for production-ready AI coding

FreeFree tier
Type
Open Source
Visit Website GitHub (74 stars)

About Advance Minimax M2 Cursor Rules

A durable execution spine for repo-scale engineering on M3 + Cursor 3.7 — with frontier-agent coding judgment and reasoning protocols distilled into rules any model can run. Tuned for MiniMax M3 (1M-token MSA context, native multimodal input) and Cursor 3.7 (Agents Window, canvases, Design Mode, /worktree, /best-of-n, Await, MCP Apps). Written to stay useful across model changes. Includes a lean always-on core, progressive depth with 18 requestable rules and 7 skill packs, M3 long-context discipline, multimodal-native support for image and video inputs, evidence-backed closeouts with explicit status labels, and portable behavior across IDEs and CLIs.

Key Features

Lean always-on core with reasoning protocol and solver loop
Frontier craft distilled (fable5-* rules) from SWE-Bench-class agents
Progressive depth: 18 requestable rules + 7 skill packs load only when needed
M3 long-context discipline: 1M-token MSA context retention and compression cadence
M3 multimodal-native: image and video input support for grounded coding
Cursor 3.7 surface guidance: Agents Window, canvases, Design Mode, /worktree, /best-of-n, Await, MCP Apps
Honest tool use with no invented tools or stale wrappers
Evidence-backed closeouts with verified/unverified/blocked status labels
Portable: same behavior in non-Cursor IDEs and CLIs via docs/AGENTS.md
Model-resilient: tuned for MiniMax M3, compatible with any Cursor-supported model

Pros & Cons

Pros
  • Lean core minimizes context overhead while maintaining rigorous execution spine
  • Progressive rule loading keeps context clean and efficient
  • M3 long-context discipline teaches effective retention and compression at 1M tokens
  • Multimodal-native support for image/video inputs grounds coding in visual evidence
  • Honest tool use ensures no invented or stale wrappers
  • Evidence-backed closeouts with verified/unverified/blocked labels increase reliability
  • Portable to non-Cursor IDEs and CLIs via AGENTS.md
  • Model-resilient design works with any Cursor-supported AI model
Cons
  • Primarily designed for MiniMax M3; performance may vary on other models
  • Requires Cursor Editor (not a standalone tool)
  • Initial setup involves copying .cursor folder into project root
  • Learning curve for users unfamiliar with Cursor 3.7 features

Best For

Repository-scale AI code engineering with structured verificationMultimodal code review and design parity using image and video inputsProduction-grade AI coding with explicit proof contracts and status labelsLong-context codebase analysis and refactoring with 1M-token support

FAQ

What is Advance Minimax M3 Cursor Rules?
It is a set of agentic-first Cursor rules providing a durable execution spine for repo-scale engineering. It includes reasoning protocols, solver loops, tool orchestration, and is tuned for MiniMax M3 and Cursor 3.7.
How do I install it?
Clone the repository and copy the .cursor folder into your project root: git clone https://github.com/madebyaris/advance-minimax-m3-cursor-rules.git ; cp -r advance-minimax-m3-cursor-rules/.cursor your-project/.cursor.
Does it work with other AI models?
Yes, it is model-resilient and compatible with any model supported by Cursor, though it is tuned for MiniMax M3 (1M-token MSA context, native multimodal input).
Is it free?
Yes, the code is open source under the MIT license.