ReadMe Launches Restless as 72% of Its Documentation Traffic Comes From AI Agents
The team behind ReadMe has launched Restless, a tool that helps APIs work better with AI agents. ReadMe reports that 72% of its documentation traffic now comes from AI agents. Restless captures live API traffic, lets agents inspect production logs through MCP, replay failed requests, and receive fix steps and documentation links inside failed responses. Agent feedback can become Linear issues or suggested documentation updates. The launch includes Agent Recovery, which embeds recovery instructions in error responses.
Reference imageEditorial responsibility
- No named human review is recorded for this page.
- Reports are grouped by semantic similarity and deterministic rules. Language models may assist titles, summaries, translation and cross-source analysis; the page reads the event directly, while its address stays stable when the title changes.
- Summary covers the current reports
Cross-source coverage
Reporting timeline
ReadMe Launches Restless as 72% of Documentation Traffic Comes From AI Agents
ReadMe reports that 72% of its documentation traffic now comes from AI agents, a striking figure given that much of the API ecosystem was designed for human developers reading docs and debugging manually. In response, the team behind ReadMe is launching Restless, a product that lets AI agents inspect real production API logs through MCP, replay failed requests, and receive recovery instructions directly inside error responses. Agents can also report broken endpoints or missing documentation, turning their own failures into suggested fixes so the API can continuously improve as more agents use it.
Read sourceReadme Says 72% of Its Docs Traffic Comes From AI Agents
Sam Altman is quoted saying "Every company is an API company now, whether they want to be or not." The Readme team, which built a tool called Restless, says roughly 72% of its own documentation traffic already comes from agents. The post argues that if a product cannot be reached that way, the agent will pick a tool that can, and that companies should treat agents as their main users. Restless is described as a tool added to an API by running one command, `npx restless init`. It records every request users' AI agents send, and when a call fails, the error response itself tells the agent how to fix it.
Read sourceRestless Launches Agent Recovery to Embed Fix Steps in Failed API Responses
Restless, the team behind README, has introduced Agent Recovery, a feature that places fix steps and documentation links directly into failed API responses so AI agents receive guidance when they encounter errors. The announcement argues that while an API error tells an agent what went wrong, what matters more is what the agent can do next. By embedding recovery instructions into failure responses, agents can work through errors rather than relying on vague promises of autonomy. The post acknowledges that not every failed call will be recoverable, but contends that helping agents resolve errors is more valuable than abstract autonomy claims. The feature targets the moment agents are stuck, aiming to improve their ability to self-correct during API interactions.
Show 2 older updatesHide older updates
Readme team launches Restless to help APIs work better with AI agents
The team behind Readme has launched Restless, a tool designed to help companies make their APIs work better with AI agents. According to the announcement, Restless captures live API traffic from agents and converts their feedback into Linear issues or suggested documentation updates. It also lets companies provide instructions on how agents should proceed after an API request fails. Failed responses can include fix steps and a direct link to documentation, giving agents a clear next step. The post frames the tool as a form of tool-use RSI.
Read sourceAI Agents Now Read API Docs and Pick Products, Analyst Warns
An X post argues that AI assistants, not humans, are increasingly the first readers of API documentation and will decide within seconds whether to use a product or a competitor's. The post cites that 72% of traffic on ReadMe's API docs now comes from AI agents. It notes that when an agent hits an error, it reads only the error message; a vague "invalid request" response becomes a dead end and the agent simply picks another tool, producing no support ticket and unexplained flat signups. The post says error messages are now a critical product surface, citing Restless for embedding fix steps and doc links in failed responses. It argues APIs were built for long-lived integration contracts, while agents need fast-changing, agent-facing layers covering settings, sign-up, and permissions. Sam is quoted saying every company is an API company now, and the post notes agents' feedback can now land as Linear issues.