June 10, 2026 — AI Gateway
Stop wiring per-provider credentials into every agent.
Set agent.astro_ai_gateway: true in astropods.yml and the deployment receives a managed gateway URL and a virtual key that covers every supported model.
Each deployment gets its own key, reused on redeploy and revoked on undeploy, and ast dev mints a short-lived key automatically so the same agent code runs locally and in production against the same gateway. Check our docs for supported models: https://docs.astropods.com/ai-gateway
The blueprint detail page now lists the agents you’ve deployed from that blueprint across every account you belong to. Scope is private to your accounts.
The sidebar’s Authors section is renamed to Contributors (it credits maintainers and commit authors, not just the original creator). The Repository row links to the repo and, when the blueprint points at a subdirectory, a second row deep-links into that subdirectory on GitHub, GitLab, or Bitbucket.
Members who linked their Slack identity now appear under their Astro identity in Insights. Spend they accumulated as a bare Slack user before linking is folded into the same row as spend tracked after linking, so each person shows up once instead of twice.
Slack users who haven’t linked still surface in the People view with their workspace profile name, avatar, and a deep link into their Slack profile.
ast agent trace is new. List traces (paginated, optionally bounded by --start / --end), pull a single trace’s detail with --trace-id, or get an activity summary with --summary — total traces, last active, and a per-day sparkline of requests and tokens. --json on any mode emits the raw response for piping into jq.ast agent logs works on any workload. The new --workload flag takes workload[/container] and resolves by exact name, suffix (vectors → my-agent-knowledge-vectors), or component label (agent, messaging, collector). ast agent get now prints each workload’s Kubernetes name next to its component so you know what to pass in.ast login preserves your active org across re-authentication. If you were working in an org when your token expired, you land back on that org instead of being bounced to personal. ast login --account <name> continues to override.ast agent subcommands now require --name or --id to identify their target. Positional names are gone — quote multi-word display names in the shell. --confirm on delete still accepts either form.ast build / ast push on Docker Desktop 4.77.0 is rebuilt on top of BuildKit’s gRPC endpoint (the same path docker buildx build uses).ast dev postgres now auto-injects POSTGRES_USER and POSTGRES_PASSWORD for container-mode postgres knowledge stores, matching the prod deploy path. Existing workarounds in agent.inputs can be deleted.PORT in the container env so frameworks that read process.env.PORT (Express, FastAPI/uvicorn, Next) bind to the right port without hardcoding.Run ast upgrade to pick up the latest CLI.
ast push / ast dev print a warning for what was skipped.ast docs agent covers the right OTel library per stack (Mastra, LangChain/LangGraph via Traceloop, Claude Agent SDK via OpenInference, raw Anthropic/OpenAI) and the env vars the platform injects automatically.Astropods publishes a Claude Code plugin with skills to migrate an existing agent, build a new one from scratch, and improve your agent card. Install from your Claude Code session:
May 28, 2026 — Know Your Agents
See consumption by agent and find out which agents people are actually using. Insights gains a By Agent / By People toggle on the Top Spenders table. The people view groups spend by member and distinguishes Unidentified (a real person from an inbound adapter who isn’t linked) from Unattributed (no person — background jobs, cron, SDK calls).
Historical traces from undeployed agents no longer inflate account totals.
For advanced agents that need control over underlying compute and data resources, you can now size an agent and give it a persistent disk at deploy time from the deploy form. CPU, memory, and storage tier are tunable, with a live cost estimate as you adjust. Data on a mounted volume survives redeploys; turning the volume off releases the disk. The deployment Metrics tab shows pod CPU and memory limits as reference lines so headroom is clear.
Editing only the advanced provisioning fields (CPU, memory, volume mount, storage size) now triggers the Save & Redeploy action like every other redeploy-relevant field.
See whether your agents are doing what they should and catch unexpected API calls before they become a bill or a security question. The agent Monitor page has a new Network Traffic section alongside Token Usage, Requests & Latency, and Traces. It surfaces inbound, outbound, and database call volume with top peers, response class breakdown, and latency.
Right-size with confidence, see exactly how much CPU and memory each agent is using before you bump its limits. The pod detail panel now has a Metrics tab with live CPU and memory charts across 1h / 6h / 24h / 7d windows.
Open the Agents page and trust what you see: live state, fresh “Update available” pills, no stale red errors. The Agents page now loads in well under a second. The “Update available” pill appears immediately after a new build is published rather than waiting on a refresh. The Launch button stays available as soon as a messaging URL exists, and the Error pill only shows when a deployment is actually in an error state (not when an old error message is still on the row).
Simpler deploy flow for self-hosted-frontend agents: no adapter picker, no validation step for something they don’t use. The deploy form now reads what an agent declares. Agents that don’t run the messaging sidecar no longer see a Slack/Web adapter picker, and the validation around it is also gone.
Reconfigure or redeploy without re-typing your API keys. Inline secrets typed directly at deploy time now round-trip through configure and redeploy. Configure shows a masked placeholder for stored values; a redeploy that doesn’t touch the secret carries it forward instead of clearing it.
Deployment endpoints now report whether each public URL is actually reachable. The web Launch button reflects this with a clear tooltip while the load balancer finishes provisioning. ast deploy --wait polls until the URL is ready, and ast agent get prints readiness alongside the URL.
Identity connections are an org-level concern now, not a personal-account afterthought. GitHub and Slack identity links now live at Settings → Connectors. Reauthorize on the GitHub connector works (previously no-op). Because GitHub’s OAuth re-auth doesn’t re-show the org selection screen for already-authorized apps, a “Request access on GitHub” link is now exposed for managing org access directly in GitHub.
esc (and ctrl+c) to cancel cleanly with a uniform footer and a Cancelled. exit.ast push and ast secrets recover from mid-command auth expiry with a single refresh + retry — no more token-expired failures during long Docker builds.ast push asks about visibility before the build step, so declining doesn’t waste build time.ast create no longer falls through to printing an empty Claude prompt when the form is cancelled.ast push accepts dev.interfaces.messaging.slack.observe_channel_ids.Run ast upgrade to pick up 0.13.2.