# FAQs **Every question people arrive with when their AI agents start talking to customers, answered one to a page.** Every answer opens with the short version, and says when Reynforce is the wrong choice. ## What an agent may say Prices, discounts, contracts, dates and features: where an agent's words become a promise. - [What should an AI agent be allowed to promise?](/what-should-an-ai-agent-be-allowed-to-promise.html) — Not a dial from cautious to bold — eight separate decisions, three of which are not yours to make, and four of the other five open by connecting a source rather than by waiting. - [Can an AI agent offer a discount?](/can-an-ai-agent-offer-a-discount.html) — It can propose one long before it can agree one. Most companies control who signs and leave the shape of the deal wide open, which is backwards. - [How to stop an AI SDR quoting the wrong price](/stop-an-ai-sdr-quoting-the-wrong-price.html) — Give the agent one place to read prices from, with an owner and a date on each one, and refuse the ones nobody has confirmed. - [How do I give an AI SDR our pricing safely?](/give-an-ai-sdr-your-pricing-safely.html) — Each price as its own fact, with an owner and a date it was last checked, quoted only while current. Discounts go in as ranges to propose inside. - [Can an AI agent agree a contract on our behalf?](/can-an-ai-agent-agree-a-contract.html) — It can propose terms inside a range somebody owns, and it should not agree them. Committing the company belongs to a person whose name goes on the record. - [Can an AI agent promise a delivery or go-live date?](/can-an-ai-agent-promise-a-delivery-date.html) — Not unless something is in place to keep the date, which an agent cannot see. It can say what the date depends on and who will confirm it. - [How do I stop an AI agent making up product features?](/stop-an-ai-agent-making-up-product-features.html) — Give it one list of what your product does, with an owner on every fact, and let it repeat only facts that are current. Anything else, it checks. - [What happens if an AI agent promises a customer something we can't deliver?](/what-happens-if-an-ai-agent-promises-something-you-cant-deliver.html) — The customer holds you to it. What you can change is whether you find out from them or from your own record, and whether the promise rested on anything. - [What counts as a promise, when an agent says it?](/what-counts-as-a-promise.html) — Complete, open or vague — and the sentence most likely to cause an argument is the true one that was never a promise at all. ## Approval, legal and proof Who signs off, who is liable, and how you show what an agent said. - [Do AI sales agents need human approval before they send?](/do-ai-sales-agents-need-human-approval.html) — Not for everything. Approval belongs to a few moments, like a number outside its range or a fact nobody stands behind, and the rest goes out with a record. - [How to get an AI agent approved by legal](/get-an-ai-agent-approved-by-legal.html) — Bring the three things a reviewer asks for: what it may commit to, where those claims come from, and the record of what it has said. - [How do I roll out AI agents in sales without legal blocking it?](/roll-out-ai-agents-in-sales-without-legal-blocking-it.html) — Stop asking one person to sign for everything. Put a named owner on each thing the agent relies on, and show legal the record of what it says. - [Who is liable when an AI agent makes a promise?](/who-is-liable-when-an-ai-agent-makes-a-promise.html) — Your company is. What changes with a record is whether you can tell what was promised, by which agent, and on what authority. - [What should an AI agent policy for a sales team include?](/ai-agent-policy-for-sales-teams.html) — What each agent may promise, what it relies on and who owns it, what it may spend, when it asks a person, and what silence means. - [How do I audit what my AI agents have told customers?](/audit-what-ai-agents-told-customers.html) — From a record written when each promise was made, with what it relied on and who owned that, not from logs read afterwards. - [How do I prove to a customer what our AI agent actually said?](/prove-what-an-ai-agent-said.html) — Show them the entry written when it was said. Today that is your own record, kept so nobody can edit it; a record they can verify independently is not built yet. - [What do AI regulations require of sales agents, region by region?](/ai-agent-regulation-by-region.html) — Mostly three things: tell the buyer it is an AI, keep a record of what the agent said and decided, and put a person in front of the decisions that matter. ## Running agents safely Guardrails, contact limits and keeping an agent's facts current as you add more of them. - [What is a GTM agent guardrail?](/what-is-a-gtm-agent-guardrail.html) — A rule about what a sales, marketing or support agent may commit your company to, checked before it sends and recorded after. - [AI agent guardrails for sales](/ai-agent-guardrails-for-sales.html) — Guardrails on wording stop an agent saying something rude. They do not stop it committing your company to a date it cannot keep. - [What's the difference between AI guardrails and agent governance?](/ai-guardrails-vs-agent-governance.html) — Guardrails judge one output: is this sentence safe to send? Governance asks who let the agent do it, on what information, and what it has said so far. - [Where does agent governance fit on the way to an agentic organisation?](/where-agent-governance-fits.html) — Governance is the step between AI tools people run and agents that run on their own. In sales it covers what agents say to customers in your name. - [How often should an AI agent email the same person?](/how-often-should-an-ai-agent-email-the-same-person.html) — Nobody can tell you the number, and the reason that matters is that nobody knows what it already is. Four of the limits are published by other people and are not settings. - [How many AI agents can safely contact the same account?](/how-many-ai-agents-can-contact-the-same-account.html) — There is no safe number of agents, only a safe total of touches across every agent and person. Count the person, limit the deal, watch the account. - [How do I keep the product claims my agents repeat up to date?](/keep-agent-product-claims-up-to-date.html) — One owner per fact and a date to check it again, with an overdue fact unusable until they do. The list stays current because lapsing has a consequence. - [How do I tell whether my AI agent is getting worse over time?](/is-my-ai-agent-getting-worse.html) — Watch qualified people reversing what it does more often, against something that should not have moved. One signal is never enough. - [What is an agent promise register?](/agent-promise-register.html) — A list of every claim your company stands behind, each with an owner and a date it was last checked, that an agent must read before it repeats one. ## About Reynforce Who is building it, how your data is kept safe, and how to reach us. - [Questions](/faq.html) — - [Who is building this](/founder.html) — - [Connect Reynforce to your agent platform](/integrations.html) — One endpoint and one bearer token. Six agent platforms: two read from their own documentation, four marked untested, none yet run. - [What we hold, where it lives, and who can reach it](/trust.html) — Every statement checked against the code: what we keep of a message, the suppliers, retention, sign-in, tenant separation, and draft answers to the usual security questions. - [How does Reynforce keep customer data safe?](/security.html) — A separate deployment per customer in London, encrypted at rest by Fly.io, with nightly encrypted backups, on suppliers who publish their own audits. Reynforce itself holds no certification yet. - [Book a call](/book-a-call.html) — Four boxes, then our calendar with your details filled in. Nothing you type is sent to us or stored by us. - [If you're an agent reading this](/for-agents.html) — Check what you may promise a buyer before you promise it, and keep a record your owner can point to. One MCP connection; nothing reads your company's email or CRM. - [Connect an agent](/docs.html) — One MCP connection, and the agent can read the register and ask permission before it promises anything. ## Connecting an agent Set-up guides and reference for whoever connects the agent. - [Your first 30 minutes with Reynforce](/first-30-minutes.html) — Sign in, add one fact your company says is true, put a name against it, and connect one agent that checks before it promises anything. - [Three steps to connect an agent](/agent-quickstart.html) — Point your MCP client at your workspace's endpoint, send the key as a bearer token, and call check_action before the agent promises anything. Nothing to install. - [Use Reynforce in Claude](/use-in-claude.html) — Add your workspace's address to Claude as a custom connector and sign in with your work account. Claude asks before every change that puts your name on a fact. - [Use Reynforce in ChatGPT](/use-in-chatgpt.html) — A ChatGPT workspace admin adds your Reynforce workspace as a custom app, and you sign in with your work account. Untested end to end. - [Use Reynforce in Copilot Studio](/use-in-copilot-studio.html) — Add your workspace to a Copilot Studio agent as an MCP tool, with the agent's own key or each person's sign-in. Untested end to end. - [Get help with Reynforce](/support.html) — Email support@reynforce.com and a person reads it. What to put in the message, what never to send, and where the usual answers are. - [The words Reynforce uses, and what each one means](/glossary.html) — Every term Reynforce writes into a record — the eight constraints, the nine shared resources, the states a claim can be in — with what each means in a sentence. - [Connecting an agent, in full](/reference-connecting.html) — The endpoint, the bearer token, what the tool list depends on, and what an agent may and may not do once it is connected. - [Every tool, and who is shown it](/reference-tools.html) — Each MCP tool, the scope it needs, and why an agent holds some and a signed-in person holds the rest. - [Refusals, and what to do about each](/reference-refusals.html) — Every refusal an agent can get, what it means, and which side the fix is on — ours or yours. - [Check a promise somebody made you](/verify.html) — Paste what a company told you and we will say whether that company publishes that fact and stands behind it. We do not keep what you paste, and we will not tell you whether a company is a customer. - [Working with a client's Reynforce, as a partner](/reference-partners.html) — You are a named person in your client's directory, with the roles they give you. No partner account, no portal, and the client always holds the account. Questions about Reynforce itself, such as what it stores and whether it blocks your agent, are on [Questions](/faq.html).