Skip to main content
The state of ai impact assessment
ED 25-03 Checklist for FedRAMP Cloud Service ProvidersFedRAMP Program
4 min readFor Cloud Service Providers (DoD/FedRAMP)

ED 25-03 Checklist for FedRAMP Cloud Service Providers

This isn't a drill. The CISA Emergency Directive 25-03 (Identify and Mitigate Potential Compromise of Cisco Devices) mandates that every FedRAMP cloud service provider act and report their status, regardless of impact level. You must complete the required actions and report to FedRAMP by 5:00 PM ET on April 29, 2026.

FedRAMP is centralizing this information to prevent agencies from sending multiple requests to each provider. Your role is to ensure this consolidation is effective. This checklist outlines the necessary steps, with a simple done/not-done test for each.

Prerequisites

Before starting, confirm you have:

  • A current inventory of every component within your FedRAMP boundary, including network and perimeter devices.
  • Access to your FedRAMP Security Inbox. The response form link was emailed there; it's not in the public notice.
  • Your FedRAMP ID (FRID) for file naming.
  • Contact details for every agency customer Authorizing Official (or ISSO) point of contact.
  • Staff with physical or data-center access, as one action requires unplugging power.
  • The FedRAMP Incident Communication Procedures ready in case you find an issue.

Checklist

1. Review CISA ED 25-03 in full.
Read the directive and its supplemental Core Dump and Hunt Instructions before touching any device. This is a required action.
Good looks like: A named owner has read both documents, and your team has agreed on scope and sequence.

2. Identify in-scope devices inside your FedRAMP boundary.
Locate all public-facing Cisco Firepower and Secure Firewall series devices supporting your cloud service offerings. Check each offering separately if you operate more than one.
Good looks like: A written list of in-scope devices per offering, reconciled against your boundary diagram and asset inventory. If you find none, document that conclusion, skip to item 8, and note that items 3 through 7 don't apply.

3. Collect logs and core dumps from affected systems.
This is recommended, as it aids the hunt. Follow CISA's Core Dump and Hunt Instructions for each device type.
Good looks like: Artifacts are collected, stored with chain-of-custody notes, and tied to specific device identifiers.

4. Evaluate every identified device for indicators of compromise.
This is mandatory. Use CISA's FIRESTARTER Backdoor Malware Analysis Report and other threat intelligence. You may submit core dumps to CISA's Malware Next Gen portal for evaluation.
Good looks like: Each device has a documented hunt result (compromise indicators found or not found) with the method and date recorded.

5. If you find compromise, anomalous behavior, or suspected federal impact, trigger incident communications.
Follow the FedRAMP Incident Communication Procedures, which include reporting to CISA and agency customers. Don’t wait until the April 29 deadline.
Good looks like: Notification goes out as soon as the condition is identified, with a record of who was notified and when.

6. If no indicators of compromise are present, apply Cisco-provided updates by 11:59 PM EST on April 24, 2026.
Apply the software updates addressing CVE-2025-20333 and CVE-2025-20362 if not already patched, plus the recently released patch for this specific persistence issue. Links by device type are in CISA's Core Dump and Hunt Instructions.
Good looks like: Every in-scope device shows the required software version, verified on the device itself, not just in a change ticket.

7. Hard reset each device by physically unplugging its power supply, no later than April 29, 2026.
A reboot isn't enough to remove the malware. Power must be physically removed.
Good looks like: Each device has a dated record of the power-pull, including who performed it, and confirmation that it returned to service patched.

8. Upload supplemental information and notify customers.
Place your file in the Incident Response folder in the FedRAMP repository, then notify all agency customer Authorizing Official (or ISSO) POCs of the completed actions. Use csv or xlsx format with the filename ED-25-03-V1-Response-[FRID]. Recommended content: the types of affected systems, a summary of actions and results (artifact collection, patching, hunting), and any additional information for customers.
Good looks like: The file is in the correct folder with the correct name, and you hold sent confirmations for every agency POC.

9. Submit the FedRAMP V1: Emergency Directive 25-03 Response Form by 5:00 PM ET April 29, 2026.
Use the link emailed to your FedRAMP Security Inbox. Providers with no in-scope systems still submit.
Good looks like: You have a submission confirmation saved to your compliance records well before the cutoff.

Common Mistakes

  • Missing the form link. It went to the Security Inbox. If nobody monitors that inbox, you may not know the form exists. The notice also states that corrective action includes public notification that a provider isn't following FedRAMP Security Inbox rules.
  • Rebooting instead of power-pulling. A software reboot doesn't satisfy the hard reset requirement.
  • Treating the patch date and the report date as the same. Patching has an earlier deadline than the status report.
  • Scoping from memory. Devices missing from your inventory are the ones most likely to be missed. Reconcile against the boundary, not recollection.
  • Skipping the hunt before patching. The patch path applies only if no indicators of compromise are present. Evaluate first.
  • Notifying FedRAMP but not agencies. Item 8 requires direct notification to each agency customer POC, in addition to the repository upload.
  • Sending nothing because you have nothing. A provider with no in-scope systems still must report status.

Next Steps

Once your response is filed, integrate what you learned back into your program. If scoping was slow, your asset inventory isn't boundary-accurate, affecting your Continuous Authorization posture beyond this directive. Document how long each step actually took so you can set realistic internal timelines for the next emergency directive. Confirm that your Security Inbox has multiple monitored recipients and an escalation path, and that your incident communication procedures specify who contacts CISA and who contacts agencies.

For questions on the directive, contact [email protected] and [email protected].

Promotional banner graphic asking if you are ready for PCI DSS 4.0 with a call-to-action to get the guide

You Might Also Like