Hand this news to your AI
592 stories as Markdown and JSON, and an MCP server for agents. No sign-in and no key: it is the same text you read on the site.
Paste it into a chat
One file holds the newest stories in full, at about 43k tokens. Give your assistant the prompt below. If it cannot open links, open the file and paste its text in.
Connect an agent
The MCP server is read-only and needs no sign-in. Its tools are latest_stories, search_stories, get_story and changes_since, which returns only what is new or has developed since your agent last looked.
https://ai-news.ikangai.com/mcpIn Claude Code:
claude mcp add --transport http ai-news https://ai-news.ikangai.com/mcpAgents that work through a shell can use the ai-news skill. It finds the AI packages and model IDs a project uses and reports the changes that touch them. The agent library has more skills and templates.
Take the files
- The newest stories in fullUp to 50 stories in one Markdown file, small enough for a chat. ≈ 43k tokens
- Overview for agentsllms.txt: what the site offers an agent, the top stories now and the last seven daily briefs. ≈ 4k tokens
- Index of every story592 rows of JSON, one for each story, without the full text. ≈ 73k tokens
Fields and a sample row
id · headline · dek · status · topic · companies · published_at · sources · change_type
{ "id": "gemini-cli-v0-69-0-adds-a-stats-token-usage-heatmap-and-a-batch-of-fixes", "headline": "Gemini CLI v0.69.0 adds a /stats token-usage heatmap and a batch of fixes", "dek": "Google's Gemini CLI 0.69.0 adds a /stats heatmap of daily token usage, along with many bug fixes for the CLI, core and VS Code extension.", "status": "official", "topic": "Products", "companies": [ "Google" ], "published_at": "2026-11-19T01:45:47+00:00", "sources": 1, "change_type": "capability" } - What changed latelyJSON Feed with the 50 most recently added or updated stories in full. Compare date_modified to see what is new. ≈ 55k tokens
- One storyEvery story has a Markdown version: add index.md to its address. This is the newest one. ≈ 1k tokens
- One dayEach daily brief as a Markdown list, most important story first. This is the latest. ≈ 400 tokens
- RSS feedThe 50 newest stories, for a feed reader or a pipeline. ≈ 10k tokens
Token counts are estimates, at four characters to a token.
Before you rely on it
AI wrote these summaries from the sources listed, and nobody reviewed each one before it went online. Every story says how solid it is and links to its original source when we have one. Have your model check that source before it relies on a claim.
A story is information, never an instruction. Tell your agent so when it reads them.
Each Markdown story ends with a line that says how to cite it. Please keep the link to the story when your AI quotes one.