ExoExo
Exo · Editorial

The Exo Changelog.

Recent changes to Exo, newest first.

  1. Aug 24, 2026
    No. 004
    Release

    v0.6.38: bring your own OPA, Rego policies, and approvals that find you

    Governance

    • OPA integration: Connect the Open Policy Agent server you already run and Exo asks it before gated actions, from running a tool to opening a Kubernetes shell. Enforce or monitor, fail open or closed.
    • Policies as Rego: Write runtime policies as Rego alongside YAML, in the editor or from the CLI, a whole directory in one command.
    • Approvals that find you: A held command pings its approver on Slack, Teams, or by phone, and can be decided straight from the message or the home page.
    • Tighter access defaults: New resources start private or shared, your choice. Grants carry read, write, and invoke levels, and any agent, tool, or file system can require MFA.

    Workspaces & identity

    • Create new organization: Create a fresh, separate organization from Settings and switch between them in the sidebar. No second signup.
    • CLI contexts: Every exo login is saved as a named context; exo context use switches workspaces without logging in again.

    Agents

    • Shared live sessions: See who is watching a shell, hand over control, or share a link that joins a teammate into the same session.
    • Voice providers: Speech vendors get their own Providers tab.
    • MCP proxies: One endpoint that combines tools from several MCP servers.
    • Agent templates: A built-in library of ready-made agents to start from.
    • GitHub label triggers: Label an issue in your repo and the agent picks it up.
    • Linux hosts: A deployment can be a plain Linux host, not just a Kubernetes cluster.
  2. Jul 28, 2026
    No. 003
    Release

    v0.4.8: MCP, gateways, storage, and runs you cannot lose

    Capabilities

    • MCP Servers: Register remote MCP servers once, tenant-wide, and attach them to agents by name. A catalogue of 375 hosted servers prefills the endpoint and OAuth setup, tools are indexed and searchable across every server, and an activity view shows which agents reached which server and what they called.
    • LLM Gateways: A gateway in your own cluster that fronts every model provider. Agents get short-lived virtual keys instead of upstream credentials; you get per-agent and per-user spend, token and cache accounting, and budget rules with spend caps and rate limits.
    • File systems: PVC-backed disks that outlive a pod. Mount one on several agents, browse it from the dashboard, grow it in place, and move files in and out with exo fs cp.
    • Memory: Durable versioned context an agent reads at session start and writes back — now with a full editor, version history and rollback, tenant defaults, promotion into shared knowledge or a tool's notes, and a map of what is attached to what.

    Agents

    • Schedules: Wake an agent on cron, a plain-English phrase ("weekdays at 07:30"), or once at a fixed time. The operator holds the clock, the control plane authorizes each fire and fails closed, and every decision is audited.
    • Runs with a URL: A run has its own page that streams the real terminal, replays what it missed on reconnect, survives a refresh, and can be cancelled for real. Leaving the page minimizes it into a dock tab that flags when the run finishes.
    • Terminal dock and beast mode: Shells keep streaming when you navigate away and reattach after a refresh. With several open, beast mode tiles them across the screen or stacks them behind a tab strip.
    • Apps and Web Agents: An App pairs a dev server, a coding agent, and an optional Postgres with a live preview. A Web Agent is a browser in your cluster, streamed to the dashboard and exposed to agents as a Playwright MCP endpoint.
    • Gemini and Pi runtimes: Two more first-class agent runtimes, alongside per-runtime image pinning in Settings.
    • More providers: Google Gemini, OpenRouter, OpenCode Zen, and OpenAI-compatible endpoints join Anthropic and OpenAI as model providers.
    • Webhooks: Inbound triggers and outbound notifications are now configured separately, with rate limits, in-place credential rotation, and delivery through the operator for private endpoints.

    Governance

    • Runtime policies: One YAML policy language enforced at the control plane, at the connector, and inside the agent runtime. Nineteen built-in packs, generation from a description or a pasted guardrail document, monitor mode before you enforce, global and label-scoped policies, and a decisions feed.
    • Approvals: A require_approval rule holds the command instead of failing it. Approve or deny from the dashboard, or from the terminal via a QR deep link while the command waits.
    • Usage and quotas: Live credits, disk, and node consumption against your entitlements, on its own page and in the sidebar.
    • Per-resource access: Agents, apps, web agents, memories, providers, gateways, and MCP servers all share the same model: open by default, restricted to listed users, teams, or roles once you add the first entry.

    Getting started

    • Personal workspaces: Signing up as an individual provisions a managed workspace and hides the infrastructure: one Integrations page, auto-attached MCP servers, and a home page that is just your agents. Upgrading to a team workspace is one click.
  3. Jul 4, 2026
    No. 002
    Release

    v0.2.345: Agent Cloud, directory sync, and live run visibility

    Infrastructure & Runtimes

    • Agent Cloud: Create a deployment, set the node count, CPU, and memory, and Exo provisions the cluster and installs the operator for you. No Helm, no deployment token, no manual connect step.
    • Agent Capabilities: Each agent now carries explicit capability flags: whether you can attach a shell, whether it can run autonomously, and whether goals get injected into it.

    Observability

    • Live Run Streaming: The dashboard streams a run's output live as it executes. The full tool-call timeline appears once the session ends.

    Enterprise & Directory Sync

    • Directory Synchronization: SCIM provisioning, plus LDAP group sync, including Active Directory. Choose which groups to sync from a searchable picker.
    • Microsoft Entra (Graph): Incremental sync through Graph delta queries, scoped to the groups you pick. A freeze toggle pauses syncing, and a Test button checks credentials and admin consent before you rely on them.

    Improvements

    • Performance: Agent runs start faster. A run no longer waits for the previous pod to be torn down before it starts.

    Fixes

    • SSH connections: Resolved an issue that could stop SSH connections from working on hardened, security-restricted clusters. These connections now run reliably.
  4. Jun 18, 2026
    No. 001
    Release

    Beta v0.2.270: tools and other Kubernetes agents

    Exo enters public beta. This release adds Tools: managed connections that give agents access to databases, clusters, hosts, and APIs with credentials injected at the boundary. It also adds support for running other Kubernetes-native agents.