# REL AI: Remediation Execution Layer

> Human-authorized, tamper-evidently audited, reversible remediation. REL plans a fix with cryptographic evidence, executes only what a mandatory three-signer authorization that includes you approves, validates the result, and reverses it automatically if validation fails. Plan-only by default. Every action is recorded in a tamper-evident, hash-chained audit ledger.

REL AI is an ElasticD3M, LLC Agent-as-a-Service product. It deploys into the customer's environment under the customer's authorization model and kill switch, in United States regions.

## What it does
1. Plan: compute the exact remediation (patch, configuration, access control, encryption, or policy) with a cryptographic digest of the intended change, and its inverse. Plan-only by default: this touches no live infrastructure.
2. Authorize: nothing that changes the environment runs without a mandatory three-signer authorization chain that includes the customer, bound to the exact plan content-hash. No auto-approve mode.
3. Remediate: apply the approved fix through a customer-authorized live connector installed at engagement.
4. Validate: re-check the target to confirm the change held.
5. Reverse: on a failed validation, automatically apply the pre-computed inverse to restore the prior state.
Every step is write-ahead logged with cryptographic chain of custody, and the customer holds a hard stop (kill switch).

## Control model
- The customer authorizes every action that changes their environment; nothing runs without sign-off; the customer holds a durable hard stop (kill switch).
- Remediation is mapped to the control it satisfies (NIST 800-171 / CMMC) for assessor-grade evidence.
- REL remediates. Network containment / isolation is its companion Eject AI, which forensically removes a contained artifact and confirms the vector is closed. REL does not itself segment or isolate networks.

## Governed Remediation Plan (self-serve, $2,500)
Buy at https://ai4rel.ai/plan. A one-time $2,500 product: submit the open compliance controls you need to close and receive a prioritized, assessor-grade remediation plan as a PDF, delivered by email in minutes. Each fix is mapped to its NIST SP 800-171 / CMMC control and objective, designed to be reversible, flagged where the blast radius is wide, and sealed with a tamper-evident content hash. Computed plan-only: it touches no live infrastructure. No calls required.

## Enterprise deployment
For in-environment remediation under your authorization model and kill switch, REL is deployed, not self-serve. To engage: hello@ai4rel.ai. Onboarding configures your authorization model, kill switch, and any live connectors before any action runs. By default REL is plan-only and touches no live infrastructure.

## Notes
Human-in-the-loop on every environment-changing action. All operations in United States regions.