AI Agent Firewall runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 5 Aug 2026.
Every agent action watched, every money-moving one gated, every verdict independently verifiable.
What the code does
We read the source, 18 h ago · rules 3dff92dd89df
Found in continuous integration, deployment or infrastructure files, or in a neighbouring package of the same monorepo. None of this is installed when you add the server: it describes how the project is built and released. We list it because a leaked key in a build pipeline is still a real problem, but it is not something this server does on your machine.
const proc = spawn(bin, [], { stdio: ["pipe", "pipe", "inherit"] });
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
Nothing serious here today
Today is the operative word: we check AI Agent Firewall every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
Alternatives to AI Agent Firewall
same job, measured the same wayGovernance layer for agentic AI — signed, verifiable receipts for every agent action.
Independent fact-checking for AI answers: a verdict and sources for every claim.
Verifiable action logs for AI agents. Every action is recorded in a SHA-256 hash chain.
Executable mathematics and independent verification for AI agents.
Verify that an agent's claimed outcome actually happened, using independently gathered evidence
Agent memory that refuses to guess: evidence-gated recall, exact-source reads, verifiable deletion.
Verifiable agent memory with signed ledger receipts — verify every write offline.
API-key pre-action risk gate: any irreversible agent action (money, SQL, delete, email).
AI Agent Firewall — questions
Answers built from our own checks of this server.