kimi-k2.5
Kimi K2.5 setup and usage patterns. Most capable subagent with 256K context and built-in vision. Use for complex reasoning and batch image analysis.
npx skills add majiayu000/claude-skill-registry --skill kimi-k2-5-opensourcesam-v2-heras-garden --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Kimi K2.5 Integration ## Overview Kimi K2.5 is the **most capable subagent** available in this project. Use it when tasks require: - Complex multi-step reasoning - Batch image/vision analysis (10+ images) - Very long context understanding (256K tokens) - Thinking mode for difficult problems --- ## Quick Setup ### Launcher Script ```powershell .\scripts\start-kimi.ps1 ``` ### Manual Environment Setup ```powershell $env:ANTHROPIC_BASE_URL = "https://api.moonshot.cn/anthropic/" $env:ANTHROPIC_API_KEY = "<KIMI_API_KEY>" $env:ANTHROPIC_MODEL = "kimi-k2.5-thinking" $env:ANTHROPIC_SMALL_FAST_MODEL = "kimi-k2-turbo-preview" ``` --- ## Available Models | Model | Use Case | Context | |-------|----------|---------| | `kimi-k2.5-thinking` | Complex reasoning, main tasks | 256K | | `kimi-k2-turbo-preview` | Fast simple tasks | 256K | **Tip:** The `-thinking` variant uses chain-of-thought reasoning, making it better for complex problems but slower for simple queries. --- ## When to Use Kimi K2.5 ### ✅ Use Kimi For: - **Complex reasoning** requiring multiple steps - **Batch vision** (10+ images to analyze) - **Long documents** (large codebase exploration) - **Difficult problems** where GLM/Mini
- Overview
- Quick Setup
- Launcher Script
- Manual Environment Setup
- Available Models
- When to Use Kimi K2.5
- ✅ Use Kimi For:
- ❌ Don't Use Kimi For:
- Vision Capabilities
- Single Image Analysis
- Batch Image Analysis
- Delegation Patterns
- Pattern 1: Fallback After GLM Fails
- Pattern 2: Long-Context Research
curl -s -X POST "https://api.moonshot.cn/anthropic/v1/messages" \
What does the kimi-k2.5 skill do?
Kimi K2.5 setup and usage patterns. Most capable subagent with 256K context and built-in vision. Use for complex reasoning and batch image analysis.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill kimi-k2-5-opensourcesam-v2-heras-garden --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.