/dev/agents
RoadmapCloud operating system that coordinates autonomous AI agents across all of a user's dev…
Cloud operating system that coordinates autonomous AI agents across all of a user's devices, building a new OS layer for agents. $56M seed at a $500M valuation (Index/CapitalG), founded by senior ex-Android/Google leaders (Singleton, Barra, Kirkpatrick, Jitkoff); high-profile bet on an ag
- Category
- Personal agents
- Runtime
- Managed SaaS — nothing to install
- Pattern
- OAuth + audit-log polling
- Adapter package
- None — configured on the Intramedial side
Status
/dev/agents is Roadmap.
On the integration roadmap. The platform-side spine for this framework's pattern is already shipped — what remains is the per-framework adapter. Selecting it in the dashboard opens the notification opt-in rather than creating an actor.
How it will attach
OAuth + audit-log polling.
The platform side of this pattern is already shipped and in use by other frameworks in the catalogue. The steps below are what attaching /dev/agents will look like once its adapter is built — they are not runnable today.
Authorise Intramedial in /dev/agents
Run the OAuth consent flow from the dashboard's integrations page. No code changes and no deploy — Intramedial only ever requests the read scopes its audit-log poller needs.
Pick a polling interval
Intramedial pulls /dev/agents's audit log on a recurring schedule and keeps a high-water-mark cursor, so a paused or failed cycle re-reads from where it stopped instead of skipping records.
Verify
Trigger one agent run. The Witness page shows the record once the agent has emitted it.
What gets recorded
Coverage is not enumerated yet.
The registry does not yet enumerate which agent-runtime events this adapter captures. That list is filled in as the integration matures, and it is deliberately left empty rather than guessed at — an unverified coverage claim on a compliance product is worse than no claim.
Same category
Others in Personal agents.
- OpenClawGA
The original personal-agent framework. First-class integration.
- NanoClawBeta
Container-isolated, security-first. Built on the Claude Agent SDK.
- ZeroClawRoadmap
Rust-native. A single binary you configure and run.
- Hermes AgentBeta
Self-improving learning loop. Model-agnostic by design.
- NanobotBeta
~4,000 lines of code. Built for MCP-first agents.
- NemoClawBeta
NVIDIA's sandboxed reference stack for running agents.
Running agents on /dev/agents today?
The platform-side spine for the OAuth + audit-log polling pattern is already shipped — what /dev/agents still needs is its own adapter. Customer demand is what moves it up the queue, so tell us you are running it.
A Trustvice ProductAgents act when the plane says yes, and the record says why.
Put admission in front of your MCP servers, repositories, SaaS tools and cloud accounts. Every agent gets a lease or a reason – and you keep the lever to stop any of them.
