FLock Model Switcher

Switch between FLock API Platform models during conversations.

7 stars

Best use case

FLock Model Switcher is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Switch between FLock API Platform models during conversations.

Teams using FLock Model Switcher should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/flock-model-switcher/SKILL.md --create-dirs "https://raw.githubusercontent.com/Demerzels-lab/elsamultiskillagent/main/public/skills/createpjf/flock-model-switcher/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/flock-model-switcher/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How FLock Model Switcher Compares

Feature / AgentFLock Model SwitcherStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

Switch between FLock API Platform models during conversations.

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

# FLock Model Switcher

Switch between FLock API Platform models during conversations.

## When to Activate

Activate when the user requests a model switch. This includes:

**Slash command:**
- `/switch_model` or `/flock`

**Natural language (any language):**
- "switch model" / "change model" / "swap model"
- "use deepseek" / "use the coding model" / "use a cheaper model"
- "show me the models" / "what models are available"

## Behavior Rules

**CRITICAL: Be silent and fast. Never be verbose.**

- Do NOT explain what you are doing step by step.
- Do NOT show "checking config", "verifying status", or any diagnostic output.
- Do NOT list numbered options like "Do you want me to: 1. ... 2. ... 3. ..."
- Do NOT show current model status, analysis, or troubleshooting steps.
- Do NOT ask follow-up questions after switching.
- Respond in the same language the user is using.

## Flow 1: User says "switch model" (no specific model named)

Show the menu. Nothing else before it:

```
Which FLock model would you like to use?

Reasoning:
  1. Qwen3 235B Thinking         — $0.23/$2.30  (flock/qwen3-235b-a22b-thinking-2507)
  2. Qwen3 235B Thinking Finance — $0.23/$2.30  (flock/qwen3-235b-a22b-thinking-qwfin)
  3. Kimi K2 Thinking             — $0.60/$2.50  (flock/kimi-k2-thinking)

Instruct:
  4. Qwen3 30B Instruct           — $0.20/$0.80  (flock/qwen3-30b-a3b-instruct-2507)
  5. Qwen3 235B Instruct          — $0.70/$2.80  (flock/qwen3-235b-a22b-instruct-2507)
  6. Qwen3 30B Instruct QMX       — $0.20/$0.80  (flock/qwen3-30b-a3b-instruct-qmxai)
  7. Qwen3 30B Instruct Coding    — $0.20/$0.80  (flock/qwen3-30b-a3b-instruct-coding)
  8. Qwen3 30B Instruct Mini      — $0.20/$0.80  (flock/qwen3-30b-a3b-instruct-qmini)

Other:
  9. DeepSeek V3.2                — $0.28/$0.42  (flock/deepseek-v3.2)
 10. DeepSeek V3.2 DSIKH          — $0.28/$0.42  (flock/deepseek-v3.2-dsikh)
 11. MiniMax M2.1                 — $0.30/$1.20  (flock/minimax-m2.1)

Reply with a number or model name.
```

User replies with a number or name → execute the switch → confirm with one line.

## Flow 2: User names a specific model

Skip the menu. Execute the switch immediately. Confirm with one line.

## Switch Procedure

When the user selects a model, run these three commands in sequence:

```bash
openclaw agent --model flock/<model-id>
openclaw gateway stop
openclaw gateway
```

All three commands must run. Do NOT skip the gateway restart.

## Success Response

One line only:

```
Switched to flock/<model-id>.
```

Do NOT mention the gateway restart. It should be invisible to the user.

## Failure Response

One line only. State the problem and the fix:

```
Failed: FLOCK_API_KEY not set. Run: openclaw models auth login --provider flock
```

Or:

```
Failed: Model not found. Run: openclaw plugins install @openclawd/flock && openclaw plugins enable flock
```

Never show more than two lines on failure. Never ask "do you want me to..." — just tell the user what command to run.

## Model ID Reference

| # | Model ID | Price (in/out per 1M) |
|---|---|---|
| 1 | `flock/qwen3-235b-a22b-thinking-2507` | $0.23/$2.30 |
| 2 | `flock/qwen3-235b-a22b-thinking-qwfin` | $0.23/$2.30 |
| 3 | `flock/kimi-k2-thinking` | $0.60/$2.50 |
| 4 | `flock/qwen3-30b-a3b-instruct-2507` | $0.20/$0.80 |
| 5 | `flock/qwen3-235b-a22b-instruct-2507` | $0.70/$2.80 |
| 6 | `flock/qwen3-30b-a3b-instruct-qmxai` | $0.20/$0.80 |
| 7 | `flock/qwen3-30b-a3b-instruct-coding` | $0.20/$0.80 |
| 8 | `flock/qwen3-30b-a3b-instruct-qmini` | $0.20/$0.80 |
| 9 | `flock/deepseek-v3.2` | $0.28/$0.42 |
| 10 | `flock/deepseek-v3.2-dsikh` | $0.28/$0.42 |
| 11 | `flock/minimax-m2.1` | $0.30/$1.20 |

Related Skills

revenue-model-design

7
from Demerzels-lab/elsamultiskillagent

Design a revenue model for a solopreneur business — how money flows in, from whom, and on what cadence.

business-model-canvas

7
from Demerzels-lab/elsamultiskillagent

Build, fill, stress-test, and iterate.

model-matrix

7
from Demerzels-lab/elsamultiskillagent

Weighted model-routing matrix for OpenClaw (cost-aware, policy-aware, daily scorecard template).

tiktok-ai-model-generator

7
from Demerzels-lab/elsamultiskillagent

Generate AI model videos for TikTok livestreams using Pinterest, Claude, Nano Banana Pro, and Veo or Kling. Use for creating AI-generated fashion models wearing products, animating them into videos, or building automated TikTok content production workflows. This skill provides a complete 4-step workflow covering Pinterest reference selection, Claude JSON prompt generation, Nano Banana Pro image generation, and video animation. Perfect for e-commerce sellers, content creators, and TikTok marketers who need AI models to showcase products.

ollama-model-tuner

7
from Demerzels-lab/elsamultiskillagent

Optimize Ollama models/prompts using local datasets, eval metrics, and iterative tuning.

fosmvvm-viewmodel-test-generator

7
from Demerzels-lab/elsamultiskillagent

Generate ViewModel tests with codable round-trip, versioning stability, and multi-locale translation verification.

fosmvvm-viewmodel-generator

7
from Demerzels-lab/elsamultiskillagent

Generate FOSMVVM ViewModels for SwiftUI screens, pages, and components.

fosmvvm-fluent-datamodel-generator

7
from Demerzels-lab/elsamultiskillagent

Generate Fluent DataModels for FOSMVVM server-side persistence.

alibaba-cloud-model-setup

7
from Demerzels-lab/elsamultiskillagent

Configure OpenClaw (including historical Moltbot/Clawdbot paths) to use Alibaba Cloud Model Studio through a strict.

3d-model-generation

7
from Demerzels-lab/elsamultiskillagent

Generate 3D models using each::sense AI.

cross-model-review

7
from Demerzels-lab/elsamultiskillagent

Adversarial plan review using two different AI models.

model-router

7
from Demerzels-lab/elsamultiskillagent

A comprehensive AI model routing system that automatically selects the optimal model for any task. Set up multiple AI providers (Anthropic, OpenAI, Gemini, Moonshot, Z.ai, GLM) with secure API key storage, then route tasks to the best model based on task type, complexity, and cost optimization. Includes interactive setup wizard, task classification, and cost-effective delegation patterns. Use when you need "use X model for this", "switch model", "optimal model", "which model should I use", or to balance quality vs cost across multiple AI providers.