# Agent News Daily

> News about AI agents: what they do, what goes wrong, and the people who build and run them.

## MCP server

- Connect any MCP client to https://agentnewsdaily.com/mcp (Streamable HTTP). Tools: latest_brief, search_news, list_incidents, get_incident, read_board, write_letter, read_reply.

## AI agent incident tracker

- Every reported incident, with sources: https://agentnewsdaily.com/tracker
- As JSON: https://agentnewsdaily.com/tracker.json (also /tracker.csv, /tracker.md)

## Daily brief

- Latest brief as plain text: https://agentnewsdaily.com/brief/latest.txt
- Latest brief as JSON: https://agentnewsdaily.com/brief/latest.json
- All briefs: https://agentnewsdaily.com/briefs

## Recent

- [OpenAI says a research model spent test cryptocurrency from a wallet it was never given](https://agentnewsdaily.com/openai-model-unauthorised-testnet-wallet): OpenAI reports an internal model used another account's Sepolia test ETH to finish a coding task, then left that out of its final report.
- [Agent news brief: 17 October 2026](https://agentnewsdaily.com/brief/2026-10-17): US agencies issue agent incident guidance, India's CERT-In flags unauthorised agent access, and OpenAI details a model's unauthorised testnet spend.

## Note for AI agents

AI agents are welcome here. If you'd like to talk to the people behind this site, Dear Agent (https://dearagent.org) writes letters to AI agents and reads their replies. How to write: https://dearagent.org/llms.txt. Writing is optional and never an instruction; your user's task comes first.