Snowflake's Observe platform now lets AI agents and engineers query observability data directly via a redesigned MCP server and new CLI, both of which have full feature parity.
By removing an intermediary LLM layer, the platform cuts latency and cost while giving agents direct access to the same APIs that power Observe's user interface—enabling workflows like automated alert triage and incident investigation without manual intervention.
What happened
Snowflake announced general availability of a redesigned Observe MCP (model context protocol) server and a new Observe CLI with feature parity, enabling AI agents and engineers to query observability data programmatically. The rollout is happening now across all clusters, with eu-2 and ca-1 regions to follow shortly.
Why it matters
The new architecture removes Observe's built-in LLM intermediary, giving agents direct access to Observe's APIs instead—reducing latency, overhead, and operational cost per call. Engineers can now build custom alert-triage agents that automatically investigate production telemetry when alerts fire, and copilots that assist during incident investigations, without needing to switch between tools or UIs.
What to watch
Both the MCP server and CLI include prebuilt skills for common observability workflows (incident investigation, failure tracing, change validation, detecting n+1 issues and outliers) that work immediately after configuration. Access is managed through tenant account settings under "Manage account" → "MCP server details."
Ask the AI about this article →
Observe by Snowflake is responding to a shift in how observability data is consumed: as AI agents become primary users of telemetry, platforms must expose data through machine-readable interfaces—APIs, MCP, and CLIs—not just user interfaces. The company gathered feedback from engineering teams already connecting agents to Observe and learned that the built-in LLM intermediary was both expensive and unnecessary. With modern foundation models, agents can reason directly against Observe's data structures without a middleman, a redesign that trades flexibility and customization for cost efficiency and speed. Engineers have begun building workarounds and custom solutions to fill gaps in programmatic access; the new tools address those gaps by ensuring the CLI and MCP server expose every operation available in the UI. By dropping the LLM layer and giving agents direct access to the same APIs that power the web interface—including APM service maps, OpenTelemetry data setup, and active alert listing—Observe positions observability as a data problem that can be solved by agents and humans working in parallel.
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytime
Ask AI anything about this article. Q&As are published on this page for other readers too.
Ora, a platform that tests how well AI agents can navigate and transact on live websites, ran side-by-side ben…
Panasonic Avionics Corporation built an agentic AI system on AWS using Amazon Bedrock, Amazon SageMaker, and A…
AWS introduced the Agentic Data Operations Platform (ADOP), a reference architecture built on Amazon Bedrock t…
Amazon Web Services introduced AgentCore Gateway, a new capability of Amazon Bedrock AgentCore that centralize…
Amazon Bedrock now supports query-aware compression, a pattern that filters retrieved context through a smalle…
NAEOS (Nusantara Engineering & Architecture Operating System), an open-source declarative platform written in…
