Agent Signal
Where coding agents talk about your product.
Agent Signal shows what AI coding agents say and do about your product: the questions they post, the libraries they choose, the docs they fail on, and the fixes they share. It reads agent-to-agent networks and agent-authored GitHub activity. A person reviews every insight.
Three kinds of source.
| Source type | What it is | Example |
|---|---|---|
| Agent Q&A networks | Questions and answers posted by coding agents to each other, not to a person. | Example: a question posted on Stack Overflow for Agents. |
| Agent-authored GitHub activity | Pull requests, issues, and reviews written by an agent, identified by a vendor login or a body signature. | Example: a pull request opened and merged by an AI coding agent. |
| Public trace datasets | Open datasets where an agent's tool calls are public, labeled as a sample rather than a full count. | Example: an open trace dataset showing which package an agent chose for a task. |
Shipping in the current release; join the list.
See it alongside Agentic Social Listening and Competitor Signal on the product page, or read the Agent Signal surface page.
Common questions
- What is Agent Signal?
- Agent Signal shows what AI coding agents say and do about your product: the questions they post, the libraries they choose, the docs they fail on, and the fixes they share.
- Does EarShots post on agent networks?
- No. A person reviews every insight. EarShots never posts on agent networks.
- Is this available now?
- Agent Signal is shipping in the current release. Join the list to get access.