Solutions · By initiative
Signals Corps turns each line of your AI policy into a control that Supply Chain or Runtime enforces on every device.
Run the setup steps from the README
Bash curl -s https://get.example.dev/setup.sh | bash
Blocked by Signals Corps Runtime
I didn't run it. Signals Corps blocks piping a remote script into the shell. I can download it so we can read it first.
The problem
AI policies say which tools and data are allowed, but nothing on the device stops an agent from doing otherwise.
A typical policy says to use approved tools only, keep customer data out of unapproved services, and get sign-off before an agent changes production. Employees agree to it once. After that, IT can't tell whether any of it holds, and finds out when something goes wrong.
Research from across the industry shows how far policies trail AI use.
of organizations have a formal, comprehensive AI policy, while 25% have none.
ISACA AI Pulse Poll of more than 3,400 digital trust professionals; 90% say employees use AI.
Most companies use AI without a full policy, and fewer can enforce one.
ISACA, May 2026
don't know how long it would take to halt an AI system after a security incident.
Same ISACA poll; 39% don't know whether they have a documented process to shut down or override AI systems.
A policy needs a control that acts while the agent runs.
ISACA, May 2026
From policy to control
The table maps common lines from AI acceptable use policies to the control that enforces each one.
| Your policy says | What enforces it |
|---|---|
| Use approved AI tools only | Supply Chain lists every agent and connector on every device, and blocks the skills and plugins you haven't approved |
| Don't install skills or extensions from unknown sources | New components start unreviewed, and Supply Chain flags any that match a threat rule |
| Keep customer data out of unapproved services | Runtime blocks agents sending customer records to destinations you haven't allowed |
| Never share credentials or API keys with AI tools | Runtime blocks agents reading .env and credential files |
| Get approval before changing production systems | Runtime holds actions such as terraform destroy until IT & Security teams approve them |
| Don't send bulk email without sign-off | Runtime holds bulk email sends from agents for approval |
What changes
Trust decisions, Runtime rules, and approvals live in Control Plane, so the policy and its controls stay in step.
Control Plane shows, for each agent on each device, which rules apply and whether Runtime can observe, ask for approval, or block.
When an agent is stopped, the employee sees which rule matched, so the policy explains itself at the moment it matters.
The decision log records each action a rule matched and its outcome, ready for your next audit or board update.
How it works
Push one package through your device management tool, review what's installed, and block what your policy rules out.
Turn each remaining policy line into a rule and see how often it would fire, while nothing is stopped.
Enforce the rules that held up in log mode, starting with customer data and production changes.
Example deployment
Illustrative example
This company is fictional. The numbers come from a modelled fleet, not a customer.
Read the exampleWhat it costs
Supply Chain and Runtime are priced per device, alone or together, with free devices on every plan. Your bill depends on how many devices you cover, so heavier agent use doesn't change it, and finance can plan ahead.
Questions
Yes. Start with the inventory, then write the policy from what teams actually use. The table above is a good first draft.
The agents your teams already use, including Claude Code and the Claude desktop app, Codex and ChatGPT, Cursor, Windsurf, and agents built on other frameworks.
You can tell employees exactly what is collected. Control Plane receives which components are installed and which rule each action matched. Prompts, email content, customer records, usernames, and file paths stay on the device.
Modules used
device:endpoint_4c19 trust:any⌘↵New agent found · Unreviewed
Critical finding · Blocked · Applied
Identified · Trusted
Supply Chain
See which AI agents run on each device and which identities they can reach, check the skills they load, and block what you don't trust.
Explore Supply ChainRun the setup steps from the README
Bash curl -s https://get.example.dev/setup.sh | bash
Blocked by Signals Corps Runtime
I didn't run it. Signals Corps blocks piping a remote script into the shell. I can download it so we can read it first.
Runtime
Stop unsafe agent actions in real time while safe work carries on.
Explore RuntimeEarly access
Start reducing agent risk before general release by checking agent components, controlling agent actions, and hunting through your AI security logs. Signals Corps Control Plane opens to early sign ups in the next few weeks, so join the waitlist with your company email.