Loading...
Haystack by deepset is a production-ready framework for building LLM applications and RAG pipelines. Jarvis SDK integrates as a custom component, giving your Haystack pipelines access to 700+ executable tools.
pip install haystack-ai requests
Sign up at jarvissdk.com and create an API key.
Build a Haystack component that calls Jarvis SDK's API.
Include the component in your Haystack pipeline.
Copy this into your project to get started immediately.
from haystack import Pipeline, component
from haystack.components.generators.chat import OpenAIChatGenerator
from haystack.dataclasses import ChatMessage
import requests, json
JARVIS_API_KEY = "jsk_your_api_key_here"
@component
class JarvisSDKTool:
"""Haystack component that executes Jarvis SDK modules."""
@component.output_types(result=str)
def run(self, module: str, action: str, params: dict | None = None):
resp = requests.post(
f"https://jarvissdk.com/api/v1/modules/{module}/execute",
headers={"x-api-key": JARVIS_API_KEY, "Content-Type": "application/json"},
json={"action": action, "params": params or {}},
)
data = resp.json()
return {"result": json.dumps(data.get("result", data))}
@component
class JarvisSDKSearch:
"""Search the Jarvis SDK marketplace for tools."""
@component.output_types(modules=list)
def run(self, query: str):
resp = requests.get(
f"https://jarvissdk.com/api/v1/catalog/search?q={query}",
headers={"x-api-key": JARVIS_API_KEY},
)
return {"modules": resp.json().get("modules", [])[:5]}
# Use in a pipeline
tool = JarvisSDKTool()
result = tool.run(module="hash-toolkit", action="sha256", params={"input": "hello haystack"})
print(result) # {"result": "...sha256 hash..."}
# Or combine with a chat pipeline
pipe = Pipeline()
pipe.add_component("llm", OpenAIChatGenerator(model="gpt-4o"))
# Route LLM tool calls to JarvisSDKTool in your pipeline logicJarvis SDK integrates as a standard Haystack @component — fits naturally into your existing pipeline architecture.
Both Haystack and Jarvis SDK are built for production. Usage-based billing, trust scoring, and error handling included.
Chain Jarvis SDK tools with other Haystack components — retrievers, generators, rankers — in any order.
Use JarvisSDKSearch to discover tools at pipeline runtime, then JarvisSDKTool to execute them.
Yes. The @component decorator is Haystack 2.x syntax. This integration requires Haystack 2.0+.
Yes. Add JarvisSDKTool as a post-processing step after retrieval — e.g., summarize retrieved documents or extract structured data.
You can define Jarvis SDK as a tool for Haystack's ChatGenerator. The component approach above gives you more flexibility for pipeline composition.
Free tier includes 1,000 executions/month. No credit card required.