# Cleric: Prevent incidents

> Agents that check every production change, then fix regressions before customers notice.
>
> Markdown version of https://cleric.ai for AI agents and LLMs. Site index: https://cleric.ai/llms.txt

Cleric checks each change you ship in production, catches the regressions your tests missed (even days later), takes the pager for your alerts, and brings you the cause, the evidence, and a fix PR to review. After the fix deploys, Cleric checks that the issue is resolved.

- Start free trial: https://cleric.ai/start
- Book a demo: https://cleric.ai/demo
- Used by engineering teams at Fubo, LaunchGood, ASAPP, Mixmax, and MRD.
- Named a Gartner® Cool Vendor in AI for SRE and Observability.

## By the numbers

- **5 min** time to root cause
- **92%** actionable findings
- **200,000+** production-grade investigations

## Ship. Move on

Once you ship, Cleric checks your release and lets you know how it went. It fixes regressions and stays on call for you.

1. **Your coding agents** plan, write, and review code before it merges. The PR merges and deploys.
2. **Change verification.** Cleric checks each change in production and fixes any regressions before they become incidents. For example, a deploy ships a bug that raises the payment retry rate while payments still go through. Cleric ships a fix PR and the retry rate returns to normal.
3. **On-call.** Cleric takes the pager, dismisses false positives, and fixes every confirmed issue. For example, an alert fires on `payment-api` 5xx errors. Cleric confirms it is a real problem, finds that a config change exhausted the connection pool, opens a fix PR for your approval, and watches the error rate until it recovers once the fix deploys.

## Stay ahead of failures

Cleric checks your releases, handles your alerts, and gets fixes ready for you to review.

### Ship without the worry

Focus on what you’re building next while Cleric checks your changes in production and catches regressions, even days later.

When a PR merges, Cleric reads what it is meant to change and what must stay the same (for example: cart load p95 does not increase, price mismatches stay at zero, database connections stay within the pool limit). It keeps checking for 14 days through weekday peaks and full business cycles, then sends you the all-clear or reports the regression.

### Give the pager to Cleric

Cleric handles your alerts and closes false positives, so you only get paged when you’re needed, with the findings ready. It works inside Slack.

When an alert does reach an engineer, it arrives with the service, why a person is needed (for example, a rollback that needs a schema decision), the cause, and the evidence attached.

### Solve complex production issues

Cleric brings you the cause, the evidence, and a fix PR to review, then checks that the issue is resolved after the fix deploys. Every line of the fix links to the production data behind it, including past investigations where the same fix worked.

## Simple on the surface. Serious underneath

Cleric draws on your stack and past investigations to find the cause, then checks that your fixes work in production.

- **Knows your stack.** Cleric maps services, dependencies, and ownership to trace regressions across systems.
- **Verifies its work.** Cleric checks each deployed fix against the live environment to confirm that the issue is resolved.
- **Compounds over time.** Cleric learns from each fix and carries that context into future investigations.

How it works in depth: https://cleric.ai/technology.md

## What customers say

> The team loves Cleric. The biggest challenge isn’t finding ways to use it, it’s deciding where to use it next. It consistently helps us uncover issues faster and shorten the path from problem to solution.
>
> — Priya Sawant, Senior Vice President of Engineering, ASAPP

> Cleric consistently delivers precise, accurate results. It’s clear the product was built by platform engineers who have lived the realities of operating production systems.
>
> — Amina Odžak, Head of Product Quality & Security, Mixmax

> No one’s going to the Cloudflare dashboard anymore. No one’s going to the Datadog dashboard anymore, even though they’re set up. The go-to is Cleric.
>
> — Affaf Ahtisham, Head of Engineering, LaunchGood

> I see Cleric functioning as an SRE companion for our Software Engineers. If one team has already solved an alert, we can apply that knowledge elsewhere.
>
> — Maxime Fouilleul, Head of Infrastructure & Operations, BlaBlaCar ([case study](https://cleric.ai/resources/case-studies/how-the-worlds-leading-community-based-travel-network-is-transforming-incident-response-with-ai))

## Paranoid by design

Read-only by default. Every action logged, every investigation auditable. SOC 2 Type II compliant, with regular manual penetration testing. Your data encrypted everywhere and never used for training.

Security and data privacy, as text: https://docs.cleric.ai/security/security-data-privacy.md

Trust Center (SOC 2 report and policies; needs JavaScript): https://trust.cleric.ai

## Built to plug right into your stack

Connect your whole stack in an afternoon. Read access by default, write access when you’re ready.

The docs list every supported integration, with setup steps and the permissions each one needs: https://docs.cleric.ai/integrations/supported-integrations.md

## Catch failures before your customers do

Put Cleric to work on your production changes.

- Start free trial: https://cleric.ai/start
- Book a demo: https://cleric.ai/demo
- Product: https://cleric.ai/product.md
- Pricing: https://cleric.ai/pricing.md