Claude visitors: new accounts receive $1.00 in bonus API credit within 30 days. One per person; usage credits only.
Use Claude in WorkBuddy.
Anthropic Claude provides Opus 4.8 for maximum capability (200K context, extended thinking), Sonnet 4.6 for balanced performance, and Haiku 4.5 for speed. WorkBuddy speaks MCP natively — register RunAPI as an MCP server once and every Claude variant becomes a callable tool inside the workbench, billed per token at 50% of Anthropic's official rate.
Use the runapi MCP server to send a chat request to Claude.
Requirements:
- Call the RunAPI chat tool exposed by the runapi MCP server
- Set model to "claude-opus-4.8"
- Ask: "Explain the difference between a mutex and a semaphore in three sentences."
- Set max_tokens to 1024
- Show me the assistant reply and the token usage counts
{
"mcpServers": {
"runapi": {
"url": "https://runapi.ai/mcp",
"transport": "streamable_http",
"headers": {
"Authorization": "Bearer $RUNAPI_API_KEY"
}
}
}
}
{
"model": "claude-opus-4.8",
"content": "A mutex is a locking mechanism that allows only one thread to access a resource at a time...",
"finish_reason": "stop",
"usage": {
"prompt_tokens": 24,
"completion_tokens": 87,
"total_tokens": 111
}
}
Use Claude in WorkBuddy in three steps
Register the RunAPI MCP server
Set RUNAPI_API_KEY in the environment that launches WorkBuddy, then add an MCP server named runapi pointing at https://runapi.ai/mcp over streamable HTTP. If you already connected RunAPI for other models, the same server covers Claude — nothing extra to install.
export RUNAPI_API_KEY=runapi_xxx
# MCP server: runapi → https://runapi.ai/mcp
Prompt the workbench agent
Type a natural-language request into WorkBuddy and name the model. WorkBuddy picks the RunAPI chat tool from the runapi MCP server, fills in model claude-opus-4.8, your messages, and max_tokens, and sends the call — no curl, no SDK code.
Ask claude-opus-4.8 via the runapi MCP server: ...
Read the tool result
The MCP tool returns the assistant message synchronously — no task polling needed. WorkBuddy renders the reply inline, and the result includes token usage counts so you can see exactly what each request cost.
usage.total_tokens: 111
Claude tool parameters on the RunAPI MCP server
| Parameter | Type | Description |
|---|---|---|
model |
string |
Required. claude-opus-4.8, claude-sonnet-4.6, claude-haiku-4.5, or any Claude variant listed in the RunAPI catalog. |
messages |
array |
Required. Array of message objects with role (system, user, assistant) and content fields. WorkBuddy builds this from your prompt. |
max_tokens |
integer |
Maximum number of tokens in the response. Defaults vary by model — set explicitly for predictable billing. |
temperature |
float |
Sampling temperature between 0 and 1. Lower values produce more deterministic output. |
top_p |
float |
Nucleus sampling cutoff. Alternative to temperature — use one or the other, not both. |
stream |
boolean |
When true, the response streams incrementally. For MCP tool calls in WorkBuddy the default synchronous result is usually all you need. |
What is Claude on WorkBuddy?
Claude is Anthropic's LLM family — Opus 4.8 with 200K context and extended thinking for deep analysis, Sonnet 4.6 for everyday coding and writing, and Haiku 4.5 for fast cheap tasks. WorkBuddy reaches all three through the single runapi MCP server, at 50% of the official Anthropic per-token rate, using the same RUNAPI_API_KEY that covers GPT, Gemini, and the image, video, and music models. Prompt caching is supported to cut costs further on repeated context.
Claude use cases
Structured reasoning over documents
Ask WorkBuddy to analyze contracts, reports, or large tables and let Claude handle the structured reasoning — the MCP tool returns clean, well-organized answers straight into the workbench.
Long document analysis with 200K context
Feed entire codebases, legal contracts, or research papers into Claude Opus 4.8's 200K token context window for analysis, summarization, and question answering without chunking.
Coding and review inside the workbench
Route code generation, pull request reviews, and multi-file refactors through Claude from WorkBuddy — Opus 4.8 for architecture decisions, Sonnet 4.6 for everyday changes.
Claude + WorkBuddy questions
Yes. WorkBuddy speaks MCP natively. Register RunAPI as an MCP server pointing at https://runapi.ai/mcp with your RUNAPI_API_KEY, and Claude appears as a callable chat tool. The same server handles chat, image, video, and music models.
RunAPI charges 50% of Anthropic's official rate. Opus 4.8 is $7.50/$37.50 per million input/output tokens through RunAPI versus $15/$75 direct. With prompt caching enabled, cached input tokens cost even less. Check the RunAPI pricing page for exact rates.
No. After the MCP server is registered, you just type a prompt into the workbench. WorkBuddy selects the RunAPI chat tool, fills in the model and messages, and shows the reply inline — no curl commands or SDK setup.
Name the variant in your prompt — claude-opus-4.8, claude-sonnet-4.6, or claude-haiku-4.5. The MCP server, config, and API key stay the same; only the model argument of the tool call changes.
Opus 4.8 for complex multi-file refactors, deep analysis, and architecture decisions. Sonnet 4.6 for everyday coding and writing — reviews, test generation, and drafting. Haiku 4.5 for fast high-volume tasks where latency matters more than depth.
WorkBuddy general setup
Not connected yet? Start with the RunAPI MCP setup guide for WorkBuddy.
WorkBuddy setup guide →Claude model catalog
See all Claude variants, per-token pricing, and context window details.
Claude models →Try Claude in WorkBuddy today.
Get a free RunAPI key, register the MCP server, and start using Claude at 50% of the official Anthropic rate without leaving the workbench.