Filings, funds, and the data behind them
Data readouts from 1.87M institutional positions across 1,824 filers, and guides to SEC filings, insider activity, and market data for AI agents. New articles daily.
Agent-Native APIs Explained: What Makes an API Agent-Ready
Most data APIs were designed for a developer reading docs. Agents need discoverable tools, metered credits, and provenance on every datapoint.
AI & Agents
12 posts
Why LLMs Get Stock Data Wrong: A Failure Taxonomy
Training cutoffs, ticker collisions, merged share classes, and invented numbers: the four ways language models botch stock data, and what fixes each.
How AI Agents Cite Sources: Auditable AI Citations
Auditable agent answers need structured citations: accession-level source links, filing dates, and UI receipts. The schema and the enforcement rules.
API Keys for AI Agents: Scoping, Metering, Rotation
Agents leak, loop, and retry. How to scope, meter, and rotate API keys when the caller is an autonomous AI agent, and why a key must never live in a prompt.
Build a 13F Tracker with AI: Raw EDGAR vs a Clean API
An honest build-vs-buy tutorial: what parsing raw EDGAR 13F filings yourself really costs, versus pointing your AI agent at a clean, provenance-tracked API.
An AI Agent Stock Research Workflow That Cites Sources
A practical end-to-end workflow for agent-driven stock research: entity resolution, 13F ownership, Form 4 insider activity, and citations back to EDGAR.
Data Provenance for AI: The Anti-Hallucination Contract
Provenance means every number an agent states carries a link to its primary source. The contract, the response schema, and the enforcement rules.
Grounding LLM Responses: RAG, Tools, or Fine-Tuning?
RAG, tool calls, and fine-tuning solve different grounding problems. For financial numbers, only source-linked tool calls make the answer auditable.
What Is LLM Tool Calling? How It Actually Works
Tool calling is how an LLM stops guessing and starts querying. Here is the schema, call, result loop, and why it beats RAG for live structured data.
MCP vs Function Calling: Protocol vs Pattern Explained
Function calling wires tools into one app. MCP publishes them once for every host. How the two layers relate, and when each fits your agent stack.
ChatGPT Market Data: How to Connect Actions, MCP, and REST
ChatGPT has no market data built in. Here are the three connection paths that actually work: custom GPT actions, MCP connectors, and REST with a bearer key.
How to Give Claude Access to SEC Filings
A practical tutorial: connect Claude to SEC filings over MCP in one command, the tools your agent gets, example prompts, and why provenance is the trust layer.