Current experiences
Try features in development and share feedback to guide their final release.Beta
As part of our beta program, you can explore newly developed product capabilities while providing feedback. Beta experiences are stable enough for everyday use but still evolving. Beta releases may be restricted to a select number of Honeycomb users (private) or open to all Honeycomb users (public). If you would like to participate in a beta, get in touch with your Honeycomb point of contact. If you are not sure who that is, Honeycomb Support can help you connect with them. No Beta experiences are currently running.Early Access (EA)
Try experimental features at the earliest stage of development and help shape them from the ground up. Our early access program provides alpha-stage features to a small group of users for feedback and validation. Early Access features may be incomplete, change significantly before release, or be removed based on what we learn. Participation is typically by invitation or request. If you’re interested in joining an early access program, get in touch with your Honeycomb point of contact. If you are not sure who that is, Honeycomb Support can help you connect with them.AI Ecosystem
Status: Early Access (EA)Type: Private
Dates: In progress AI Ecosystem brings together the views you use to monitor and investigate your AI agents: fleet-wide performance, estimated LLM cost, and the conversations behind them. Check the health of your agent fleet, find what drives your LLM spend, and drill down from a trend to the conversation that explains it in the Agent Timeline. To learn more, visit What is AI Ecosystem?.
Timeline Insights
Status: Early Access (EA)Type: Private
Dates: In progress Timeline Insights automatically surfaces findings, such as failures, retry loops, latency, and token cost, when a conversation loads in the Agent Timeline. Review the most severe issues in a conversation without reading through every step. Timeline Insights requires Honeycomb Intelligence to be enabled on your team. To learn more, visit Agent Timeline: Timeline Insights.
Past experiences
Explore which features we’ve tested in the past and how they shaped our product today.Beta
An archive of our past beta programs. These features were nearly production-ready and were refined based on user feedback before graduating to general availability.Canvas Connectors
Type: PublicDates: Graduated to general availability Connectors bring the services you already use into Canvas. Once you connect a service, the Canvas Agent can use it during an investigation alongside your Honeycomb telemetry. For example, the Canvas Agent can review an incident.io timeline, check who is on-call in PagerDuty, or read your runbooks in Notion. To learn more, visit Canvas Connectors.
Anomaly Detection
Anomaly Detection represents a fundamentally different philosophy in Honeycomb. Rather than requiring teams to predict what might go wrong and configure alerts accordingly, Honeycomb learns your service patterns and proactively notifies you when behavior deviates from normal. To learn more, visit Anomaly Detection.Activity Logs
Type: PublicDates: Graduated to general availability Activity Logs give you visibility into changes made to key resources in your environment, whether by a person or an automated process. To learn more, visit Investigate Team Activity.
Chat about a page
Type: PublicDates: Graduated to general availability Chat about this page lets you ask questions about the Honeycomb page you are on in natural language. To learn more, visit Chat About a Page.
Honeycomb Canvas Slack app
Type: PublicDates: Graduated to general availability Use Honeycomb Canvas directly in Slack to ask questions of your data, explore your system, and investigate problems without switching contexts. Collaborate with your teammates by mentioning
@Honeycomb Canvas in any channel.
To learn more, visit Ask Canvas in Slack.
Metrics
Type: PrivateDates: Graduated to general availability Metrics is a major update to Honeycomb’s legacy metrics experience, built to align with the OpenTelemetry specification for metric types and behaviors. This upgrade introduces a more expressive query model, advanced temporal functions, and full support for alerting via Triggers. To learn more, visit Metrics in Honeycomb.
Model Context Protocol (MCP)
Type: PublicDates: Graduated to general availability The Honeycomb Model Context Protocol (MCP) enables AI assistants to directly interact with your Honeycomb observability data through natural language. This open standard allows seamless communication between AI models and Honeycomb, transforming how you investigate issues and analyze system behavior. To learn more, visit Honeycomb Model Context Protocol (MCP).
Text panels on Boards
Type: PublicDates: Graduated to general availability Text Panels let you add narrative context directly to a board, alongside your queries. Use them to explain what the data means, highlight key insights, or link to runbooks and playbooks so your team knows what action to take. To learn more, visit What is a Board?: Panels.