
Every rule in one place.
Scoped to each agent and workflow. Update one, not your whole prompt.
Infrastructure to control what your agents are told, and when.
It was buried in a long prompt, or sitting in a skill the agent never opened.
You can write the rule. You can’t control whether your agent sees it when it matters.
Say what your agent should do, in plain language.
Preview the past conversations it would have reached.
Your agent gets the rule at that step of the run, and you see which conversations it reached.
Compare outcomes with and without the rule, and see which mistakes still repeat.
import { Mirrors } from "mirrors-sdk";
const mirrors = new Mirrors({ apiUrl, apiKey, agentId });
// Once per user turn: the rules that apply right now.
const { context } = await mirrors.getContext({ trace });
const system = context ? `${prompt}\n\n${context}` : prompt;
// Your model call, tools and streaming stay as they are.
mirrors.captureTrace({ trace, final: true });
Scoped to each agent and workflow. Update one, not your whole prompt.

Mirrors delivers it only where you said it applies.

Compare outcomes with and without each rule.

1,000 turns a month, free
INCLUDES

$0.005 /turn
EVERYTHING IN HOBBY, PLUS

Volume rates on a usage commitment
INCLUDES
| Hobby | GrowthMost teams | Enterprise | |
|---|---|---|---|
| PRICE | |||
| Base fee | $0 | $299 a month | Custom |
| Turns | 1,000 a month, free, shared across all your agents | $0.005 per turn | Volume rates on a commitment |
| Agents | Unlimited, no agent fees | Unlimited, no agent fees | Unlimited, no agent fees |
| Rules | No charge | No charge | No charge |
| MCP reads and rule lookups | No charge | No charge | No charge |
| Rate guarantee | Does not apply | 6 months from your first payment | Set in your contract |
| BILLING | |||
| Billing | None | Monthly. Base fee at the start of each period, usage at the end | Set in your contract |
| Turns billed | Does not apply | Every turn counted, never rounded up | Every turn counted |
| Usage dashboard and estimated charges | Does not apply | Included | Included |
| Invoices and payment methods | Does not apply | Included | Included |
| Cancellation | Does not apply | Anytime, effective at the end of the period | Set in your contract |
| PRODUCT | |||
| Find recurring mistakes | Included | Included | Included |
| Write and preview rules | Included | Included | Included |
| Publish rules and deliver them at runtime | Included | Included | Included |
| SDK, API and MCP access | Included | Included | Included |
| SUPPORT | |||
| Founder onboarding | Not included | 30-minute call and first-workflow review | Included |
| Integration help | Not included | Included | Included |
| Support | 24/7, direct with the founders | Terms agreed in writing | |
| Larger implementation work | Not included | Scoped and quoted separately | Scoped and quoted separately |
A turn is one user message and your agent’s full response, tool calls and subagents included. Retries are never billed.
Full pricing detailsMirrors is the guidance system for production AI agents. You write a rule once and define where it applies. Mirrors delivers it to your agent at that step of the run, shows which conversations it reached, and measures whether it helped. It also surfaces the mistakes your agent keeps making, so you know which rule to write next.
A rule in the system prompt rides along on every request, so rules pile up and each one gets buried among the rest. A skill is read only when the model decides to open it, so the rule can be missed. Mirrors delivers each rule where you said it applies, keeps every rule in one place, scoped to the right accounts and workflows, and shows whether it helped.
No. Mirrors makes sure the rule reaches your agent where you said it applies, and shows you the conversations it reached. Whether the model then follows it is what the measurement is for: you compare outcomes with and without the rule on real runs.
Observability shows what your agent did. Evals check it against test cases before you ship. Mirrors controls what your agent is told in production: it delivers your rules where they apply, shows which conversations each one reached, and measures each one on real runs. It works alongside both.
Install the Mirrors SDK, send your agent’s conversations, and request the rules that apply once per user turn. It works with any agent framework and any model, and your app keeps control of its model calls and tools.
Mirrors answers in 200 ms on average, once per user turn. It fails open: if a response takes longer than 750 ms, your agent continues without the rules.
Yes. Mirrors previews each rule against your stored conversations, so you see every one it would have reached, including any you didn’t expect, before you turn it on.
Every plan includes unlimited agents. Hobby is free with 1,000 turns a month shared across your agents, and you can sign up without talking to anyone. Growth is $299 a month plus $0.005 per turn. Enterprise is custom pricing for high volume.
Start free with 1,000 turns, or book a demo.
Start free