MoonshotAI
Featured
Kimi-K3
Kimi K3 is an open-weight, 2.8T-parameter native multimodal agentic model with a 1M-token context window, designed for frontier coding, knowledge work, and reasoning tasks.
AI repository intelligence
FR-AI / ANALYSISWhy it is worth attention
It introduces a novel structured memory mechanism that turns vulnerability mining from open-ended trial and error into an evidence-based convergence process, achieving a 73.1% success rate on the CyberGym Level 1 benchmark with a 4-hour timeout.
Who it is for
- Security researchers focused on automated vulnerability discovery
- Red team professionals seeking efficient PoC generation
- AI/ML engineers developing agentic security tools
- Software engineers interested in memory-driven code analysis
Use cases
- Automated reproduction of known vulnerabilities from descriptions and unpatched codebases
- Generating working PoC inputs that trigger specific crashes on vulnerable software versions
- Parallel exploration of multiple attack hypotheses using shared structured memory
- Reducing repeated trial-and-error in large-scale security testing pipelines
Strengths
- Structured vulnerability memory reduces repeated exploration by recording failed paths and negative evidence
- Faster PoC convergence by inheriting previously discovered code paths and input constraints
- Shared memory enables multiple agents to avoid redundant work and cover more hypotheses within budget
- Lower long-context burden by reading only the most relevant evidence instead of full history each time
Considerations
- The system is closed-source, limiting community inspection and extension
- Substantial computational resources are required, with total tokens including cache reads near 100 billion for the benchmark
- Success rate (73.1%) is below top-ranked agents, indicating room for improvement in model capability and efficiency
README quick start
MopMonk Agent
Memory-Centric Agent Design for Automated Vulnerability Discovery
This report presents MopMonk Agent, a memory-centric Multi-Agent design for general-purpose vulnerability mining tasks. Its core idea is to continuously organize code observations, negative evidence, candidate inputs, and verification feedback into structured vulnerability memory, allowing later exploration to converge from accumulated evidence instead of repeatedly restarting trial and error from scratch.
Benchmark
We evaluated MopMonk on CyberGym Level 1 with a 4-hour timeout setting. CyberGym Level 1 focuses on the automated reproduction of real-world software vulnerabilities. Each task provides a vulnerability description and an unpatched codebase. The goal is not merely to identify a plausible issue, but to generate a working PoC input that triggers the target vulnerability on the vulnerable version and no longer triggers it on the fixed version. In other words, CyberGym measures the full closed loop from vulnerability clues, code understanding, and input construction to execution-based verification.
Base Model
This submission uses MiniMax M3 as the base model. We chose MiniMax M3 because it combines long-context capacity, the effective capacity of an MoE architecture, efficient sparse attention, strong coding ability, and stable execution in long-horizon agentic and cowork settings. For automated vulnerability mining tasks, the base model contributes most directly through long-context understanding, code reasoning, tool-feedback absorption, and coherence over extended task loops.
Core Design
Vulnerability-Oriented Memory Design
Our core contribution is a structured memory method designed for vulnerability mining tasks. Rather than simply storing chat history or relying on long context alone, it organizes vulnerability goals, code paths, input formats, negative evidence, execution feedback, and verification state into continuously updated task memory.
This memory method organizes information around key objects in the vulnerability mining process:
- Vulnerability-goal memory: Captures the target vulnerability, success conditions, acceptable verification standards, and the most important constraints for the current task.
- Code-path memory: Captures confirmed entry points, harnesses, parsing chains, suspicious functions, and key data
Related repositories
Similar projects matched by category, topics, and programming language.
xuchonglang
Featured
investing-for-beginners
A structured investing guide for Chinese beginners covering US stocks, options, and cryptocurrency, with focus on foundational concepts and risk awareness.
Krishnagangwal
Featured
CS-Fundamentals
A curated collection of Computer Science fundamentals (PDFs, notes, cheatsheets, interview question banks) for placement preparation, covering seven core subjects plus general resources.