Observe
Observe
The Observe section of the dashboard provides visibility into how AI is used across the organization. It answers questions like: what is AI spend and where does it go, which MCP servers and tools are used most, what did an agent actually do in a session, and who in the organization has adopted the platform.
All Observe pages share a common filter bar (date range, user, agent, account type) and an AI-assistant mode that answers natural-language questions over the same data. Observe pages require the organization admin role.
Sections
Section titled “Sections”Track AI spend across the organization. Headline totals for tokens, cost, active users, and AI clients, with breakdowns by role, employee, model, agent, and more. Drill from org-wide totals down to individual sessions, compare periods, and export session-level data.
Usage analytics for MCP servers and tools: success rates, event volumes, active users and targets, activity by source, skill usage, and error trends. Use it to see which tools deliver value and where failures cluster.
A searchable list of agent chat sessions with risk indicators. Open any session to review the full transcript, including tool calls, token usage, duration, and flagged risk findings.
The raw stream of tool executions across every MCP server, skill, and local tool. Filter by server, user, status, or arbitrary log attributes, expand traces, and inspect the full input and output of any tool call.
Adoption tracking for the organization: who is enrolled, who isn’t, and how actively each member uses AI. Drill into any member for their tools, models, costs, sessions, and risk events.
Getting data into Observe
Section titled “Getting data into Observe”Observe pages populate once the observability plugin is installed in the AI agents used across the organization (Claude, Cursor, and Codex are supported from the setup dialog). Until then, pages show a setup prompt with per-provider installation steps.