Does Cornerstones place trades?
No. Cornerstones provides market context; trading authority stays in the customer runtime.
Agent-native market data
Use Cornerstones as a read-only market context layer. The LLM can inspect permitted data and evidence, while order routing, risk controls, and portfolio authority stay in the customer runtime.
How should LLM trading workflows get current context before analysis without granting execution authority?
Separate read-only market context from execution and let the customer trading system own risk, sizing, approval, and order placement.
cornerstones-client verify
# request read-only context, then run trading analysis in your own runtime| Option | Best for | Tradeoff |
|---|---|---|
| Cornerstones | Agent-native market context | Requires adopting a product-level access model |
| Raw private integration | Human-operated internal systems | Can expose credentials, adapters, and unstable implementation details to agents |
| Static prompt paste | One-off prototypes | No freshness, usage accounting, entitlement boundary, or repeatable evidence trail |
No. Cornerstones provides market context; trading authority stays in the customer runtime.
It lets agents reason with current information without giving them direct execution tools.
Yes, when the account plan permits charts, options, macro, orderflow, or deeper liquidity context.
Log the market context used, freshness signals, and plan boundary so later reviews can reconstruct the analysis.