SpecStream for Haystack

What you get

• Columns as a fields dictEach Document's meta carries source_url, the time the row was read, and its columns as a fields dict for the next component.
• Saves with the pipelineA pipeline saved to YAML keeps the fetcher's max_price_usd and stores only the name of the env var that holds the key.
• Reruns reach the same runA retried step with the same request and limits reaches the run it started that hour, and its rows are paid for once.

Set it up

1. 1Run pip install -U specstream-haystack on Python 3.10 or later, with haystack-ai 2.14 or newer.
2. 2Set SPECSTREAM_API_KEY to a key from Agents & API; the components read it through Secret.from_env_var.
3. 3Add SpecStreamListFetcher(max_price_usd=3) to a pipeline, or give SpecStreamListTool(max_price_usd=5) to a Haystack Agent.

The full setup guide has every step, with what to do if something doesn’t arrive.

• OpenAI APILive web records as an MCP tool in the Responses API
• LangChainAgent tools and a retriever that returns LangChain Documents
• LlamaIndexA tool spec and retriever nodes for LlamaIndex agents
• CrewAILive web rows for the agents in your crew
• VercelAI SDK tools in TypeScript, plus a client for route handlers
• BotpressComing soonHub actions that put live web rows in a bot's reply