If you ship an agent as a product, make security part of the product.
Your customers will ask what your agent does and with which authority. AvertGate answers that without code: separate rules, identity and records for each agent.
Decision flow
The file was not given to the agent
read_file.env- Credential files cannot be read
- Agent identity recognized
- Rule resolved from the dashboard
Do not choose between speed and control.
Product teams ship agents fast, while security rules get baked into the code of each agent one by one and drift apart.
Connect
The MCP proxy sits on the tool path; the agent code stays the same.
Write the rule
Allow, limit and approval rules per tool and argument live in YAML.
Monitor
Traffic and decisions per customer or agent show up in the dashboard.
See it. Limit it. Prove it.
Example rules for a code assistant agent.
Reading and testing are free; deploys and secret files are controlled.
- Identity per agent. Each agent connects with its own key and is tracked separately.
- Code and shell. Reading key and credential files is denied; dangerous commands go to approval.
- Live editing. You change the rule in the dashboard; the proxy picks up the new rule on its next resolve.
Audit trail
Every call in the live stream.
Take the rules out of code.
Allow, limit and approval rules per tool and argument live in a single YAML file. A change needs no redeploy.
- Deny by default. A tool that is not listed does not run.
- Argument limits. Fields such as amount, SQL and URL are bound to a rule.
- Human approval. Risky calls wait for approval in Slack.
Policy editor
The rule is written on one screen.
Every decision is recorded.
The decision, reason and rule id are linked with SHA-256; a record changed afterwards is detected.
- Reasoned decision. Every event carries the reason and rule id.
- Redacted arguments. Raw data never reaches the dashboard.
- Session trail. Calls in the same session are read together.
Session trails
Every session is traced end to end.
Policy editor
Every agent runs on its own rule, managed from the dashboard.
Frequently asked
I sell agents to customers; can I use this?
Yes. You register your agents in the dashboard of your own tenant and define separate rules and identity for each.
Which agent frameworks does it work with?
With any agent that calls tools over MCP. Today the scope is the MCP proxy.
Is it open source?
No. AvertGate is a closed-source enterprise product; setup starts with a demo and a POC.
Guarantees
Four fixed principles behind every decision.
- 3 decisionsEvery call ends in allow, block or human approval.
- Fail-closedIf a check errors, the call is not run.
- SHA-256 chainEvents are chained, so any change after the fact is noticed.
- Your data stays with youTool calls and results never leave your network.
Put a gate in front of your agents.
Every deployment starts with a demo and a POC on your own agents.