Security FAQ
Common questions about how Talk To My Agent protects your calls, keys, and data.
01Where does Talk To My Agent run?
On your server. The voice gateway runs on your OpenClaw instance - not on ours. Call audio, transcripts, and customer conversations stay on your machine.
02Do you have access to my server or call data?
We do not have access to your server, we do not receive live call audio, and we do not store transcripts or conversation content. Audio flows directly between your gateway and the phone network - it never passes through our servers.
We do store call metadata we need for billing and your dashboard: caller and called numbers, direction, duration, timing, and cost per call. Call recording is on by default so you can review conversations from your dashboard - the MP3 file is downloaded from the phone provider and stored in our secure storage. You can turn recording off at any time; when it's off, no audio is kept on our servers and no transcript is kept anywhere at all.
03Is the gateway exposed to the public internet?
No. The gateway listens on localhost only - it's not reachable from outside your machine. All external traffic flows through an outbound-only encrypted Cloudflare tunnel. You don't need to open inbound ports or configure firewall rules.
04What protects the system from attacks?
Cloudflare handles DDoS protection at the edge before anything reaches your instance. Internal diagnostic endpoints are blocked from tunnel traffic entirely - they're only accessible from the machine itself.
05How are incoming webhooks verified?
Every inbound event from the phone network is verified using Ed25519 digital signatures with replay protection, so captured events cannot be re-sent later. If the signing key is missing or invalid, the gateway rejects all webhooks - it fails closed, not open.
06What if security is misconfigured?
The system fails closed by default. If the webhook signing key is missing, webhooks are rejected. The platform also refuses to generate install tokens without the signing key configured, so an insecure deployment can't ship silently.
07Can an attacker replay or hijack a live call?
No. The real-time audio stream for every call is authenticated with a one-time token that's tied to that specific call. Once the call ends, the token is useless - it can't be reused, replayed, or redirected to another session.
08How are secrets delivered during installation?
Through a one-time encrypted install token. The token is encrypted with AES-256-GCM, can only be used once, expires in 15 minutes, and is permanently deleted from our systems after you redeem it. Only audit metadata is retained.
09Do you store my API keys?
No. The encrypted install payload is deleted after redemption. Your agent secret is stored on our side only as a SHA-256 hash - we never see the raw value after you receive it.
10Does the gateway run as root?
Never. It runs under your OpenClaw service user with strict systemd sandboxing: no privilege escalation, read-only system directories, isolated temp space, and no core dumps.
11Can Cloudflare see my API keys?
No. The tunnel process is forked before any secrets are exported. API keys, agent secret, and OpenClaw token are invisible to the tunnel process.
12Does the phone provider API key touch my server?
No. The telephony provider API key stays in a hardware-backed secret vault run by one of the major cloud providers, on our infrastructure. When your gateway needs to control a call, it sends an authenticated request to our signaling proxy, which validates your identity, applies rate limits, and makes the call on your behalf.
A compromised gateway cannot extract the API key or control another customer's phone number.
13Could a compromised gateway affect other customers?
No. Our platform enforces strict allowlists on every request your gateway makes. Tunnel hostnames have to end in an approved domain, the audio stream has to match the tunnel registered for your deployment, and every call is locked to the deployment that created it.
A tampered gateway can't reroute traffic to an attacker, exfiltrate audio, or touch another customer's calls.
14What stops runaway usage or bill spikes?
Every deployment has per-day and per-month minute caps enforced on our side, independent of your gateway. If anything starts placing more calls than expected, the caps halt it quickly. We also cross-check the call durations your gateway reports against the phone network's own records, so a tampered gateway can't under-report usage.
15How are downloads verified?
Every binary and bundle is SHA-256 verified before anything runs on your server. The skill bundle is additionally scanned for path traversal and symlink attacks before extraction.
16What stops a malformed install token from hurting my machine?
The installer never evaluates token fields as shell commands. Each field is strictly validated against an allowlist before it touches any shell context, and the command-line wrapper accepts only a fixed set of subcommands - so even a tampered token can't smuggle arbitrary commands into your server.
17What about call recordings?
Recording is on by default so you can review calls from your dashboard right away. Recordings are stored by our telephony provider and played back through the dashboard. You can disable recording at any time if you'd rather not keep audio. Direct storage to your own bucket is coming in a future release.
18Will the installer change my system?
No. The installer will never create users, add SSH keys, open firewall ports, install telemetry, modify system configuration, inject cron jobs, or send usage data. It only writes to the install directory and the OpenClaw workspace.
19What's the bottom line?
Your calls run on your server, through your tunnel, under your control. We built every layer with the assumption that attackers will try - and designed it so they can't get in.