Roadmap
Roadmap: The AstraDot Agent-Ops Toolkit
We are building independent tooling for teams running always-on agents. Here is what is live and what is coming.
Last verified: September 30, 2026 · Reviewed by AstraDot editorial desk
This is an independent project, not affiliated with OpenAI. It exists because always-on agents like Dots change the operating questions for a team — and we would rather answer those questions in the open than guess.
- Now — liveLive
The independent guide
Source-checked explainers on OpenAI Dots and GPT-6 Astra, with every page stamped with its review date and unknowns marked as not published.
- Now — liveLive
Free evaluation tools
The Agent Readiness Scorecard and the Dot cost comparator — both client-side, both honest about unpublished figures.
- Next — private betaBuilding
Agent-ops toolkit
Practical tooling for teams running always-on agents: workflow triage, an allowance and budget tracker, and a governance policy builder (Custom Rules, approvals, activity review).
- Later — plannedPlanned
Team workspace
A shared space to register agents, assign owners, and keep an audit trail of what each agent can do and who approved it.
- Later — exploringPlanned
Integrations
Connectors and exports so the toolkit fits the systems teams already use. Nothing here is committed yet.
How we decide what to build
- Start from a real operating question teams hit with always-on agents.
- Prefer honest "not published" over invented precision.
- Keep humans in control: approvals, audit trails, clear ownership.
- Build in public and let the waitlist shape priorities.
Want in? Join the waitlist to get early access and influence what ships first.