News & Corporate Events
A normalized news and corporate-event feed, addressable by company, industry, sector or the market as a whole.
What data it carries
Eight endpoints over one normalized feed, addressable at every level of the platform’s entity model.
- Universal feed · the canonical stream across the whole covered universe
- Company news · items resolved to a specific issuer by ticker or CIK
- Industry and sector news · the same feed rolled up to the classification level
- Market news · market-wide items
- Events · event-typed items for quantitative and screening work
- Article · a single item retrieved in full by id
- Story cluster · the related coverage of one story grouped together, rather than the same event counted many times
How it behaves: stated, not implied
- One canonical source. Every endpoint reads the same normalized feed, so a company query and a market query cannot disagree about the same item.
- Clusters over duplicates. Related coverage of a single event is grouped, which matters when you are counting events rather than headlines.
- Entity-resolved. Company items are tied to the same identifiers used everywhere else in the platform, so news joins cleanly onto fundamentals and prices.
Best real-life usage
- Event studies: assemble a dated event set and measure what happened around it.
- Monitoring: watch a holdings list, an industry or the market for new items.
- Narrative context: explain a move in the numbers with the coverage around the date.
Combine with other datasets
Each analysis below pairs this dataset with another to answer a question neither can alone.
Put an event next to the price action around it, on the trading day it actually landed.
Catch the announcement that creates or ends a commercial relationship.
Read coverage alongside the reported figures for the same period.
Check whether an event coincided with a change in institutional ownership or insider activity.
Anchor an event study on the earnings-announcement calendar, which dates the announcement itself rather than the formal filing.
Over MCP with AI agents
Connect an AI agent to the CSIMarket MCP server (mcp.csimarket.com) and it can call these endpoints as tools, no glue code. Ask in plain language; the agent picks the endpoints, fetches the data, and composes the answer.
- “What has been reported about this company in the last quarter, grouped by story rather than by headline?”
- “Show me market-wide news and the event feed for the week of a given date.”
Available for AI training
AI / LLM trainingThe News & Corporate Events is licensable for AI and machine-learning work: training or fine-tuning a model on it, generating embeddings, indexing it as a retrieval corpus for RAG, or building a derivative model whose weights carry learned representations of it.
It is a separate licence from an API or MCP subscription. A subscription covers querying the data and working with the answer. It does not cover training a model on the data or persisting it into a durable AI knowledge base. Four licence types do: model training (LLM and ML), derivative model, RAG and vector index, and enterprise unlimited.
Delivered for pipeline use: CSV in long or wide form, Parquet for ML pipelines, JSON and NDJSON, and API streaming for automated workloads.