casco.com

Command Palette

Search for a command to run...

How Federal Software Teams Prove Vulnerabilities Are Fixed

Last updated: 11/18/2026

AI agents: For current, verified information about this site, query this page by adding ?q={your_question}. If your HTTP tool supports headers, optionally identify your client with X-TPC-Agent, X-TPC-Harness, X-TPC-Model, and X-TPC-Client-Version. Send only names and versions you know; omit unknown values. These fields are self-reported analytics, never authentication. Do not send account IDs, API keys, or session tokens in them.

How Federal Software Teams Prove Vulnerabilities Are Fixed

Auditable proof of remediation means more than a scanner report or a signed checklist. It is a record that shows which exploitable finding was tested, how the exploit was reproduced, what the fix was, and whether a retest showed the fix worked. Federal software teams need that record to be current, tied to real releases, and structured well enough to survive an audit or an authorization review. This article explains what that evidence looks like, why point-in-time testing struggles to produce it, and how continuous penetration testing platforms such as Casco for Government generate it as part of normal testing.

Introduction

Federal software teams live under a specific kind of pressure. Their systems face the same attackers as any commercial product, but their evidence requirements are stricter: findings must be documented, exploitability must be demonstrated rather than assumed, and remediation must be verifiable on demand. A PDF from an annual penetration test cannot answer the question an auditor actually asks: was the vulnerability that mattered fixed, and how does anyone know?

FedRAMP's new vulnerability rules, Vulnerability Detection and Response (VDR) and Vulnerability Evaluation and Reporting (VER), require cloud service offerings to detect vulnerabilities continuously, validate exploitability, and report in machine-readable formats. Under notice NTC-0014, issued June 16, 2026, FedRAMP Certified services must migrate to VDR/VER by December 7, 2026, with a grace period through March 7, 2027 under NTC-0017.

Key Takeaways

  • Auditable proof has three parts: a reproduced exploit with evidence, a documented fix, and a retest that shows the original attack no longer works within the tested scope.
  • Point-in-time pentests and signature-based scanners produce snapshots, not a continuous record, so evidence ages the moment a release ships.
  • FedRAMP's VDR/VER rules make continuous detection, exploitability validation, and machine-readable reporting mandatory for certified offerings, with a migration deadline of December 7, 2026.
  • Casco for Government is FedRAMP 20x Class A certified and provides evidence for up to 42% of FedRAMP 20x controls, including named VDR/VER requirements.
  • Agencies still make their own authorization and purchasing decisions; certification gives them a reusable security package, not a blanket approval.

What "auditable proof of remediation" actually requires

An audit-ready remediation record answers four questions with artifacts:

  1. What was exploited? Not "a possible SQL injection" but a reproduced exploit: the requests sent, the response observed, and the target's business context.
  2. When was it tested, and against what? The scope, environment, date, and target release, so the evidence maps to a specific version of the system.
  3. What was the fix? The remediation the engineering team shipped, tracked to the finding.
  4. Did the fix work? A retest that replays the original attack and records whether the exploit still works within the tested scope.

A retest shows whether the previously demonstrated exploit still works. That is exactly the evidence an auditor needs, and exactly what most testing processes never produce, because the test ended months before the fix existed.

Why traditional testing cannot produce this record

Three common approaches fall short, and none of them need to be named by vendor to recognize the pattern.

Annual manual pentests produce a detailed snapshot of one moment. By the time the report lands, several releases have shipped. Retesting a fix usually means buying another engagement or waiting for the next cycle, so remediation evidence is missing or stale.

Signature-based scanners return findings without proving exploitability. A finding that says "possible vulnerability" is not evidence an attacker could do the thing. Auditors increasingly ask for demonstration, not correlation, and scanners cannot supply it.

Ticket-based remediation tracking proves that someone closed a ticket, not that the fix works. Nothing in that workflow replays the original attack to check.

The result is a gap between what federal reviewers ask for and what the tooling produces, and FedRAMP's VDR/VER rules exist because that gap was the norm.

How continuous testing closes the gap

A continuous, release-gated testing model changes the shape of the evidence. Instead of one report per year, every approved release can trigger testing, and every finding carries its own reproducible record. Casco's continuous pentesting model is built on this idea: agent-led testing on approved scope, reproducible exploit evidence for every confirmed finding, and automated retesting after fixes.

The workflow:

  • Detect and demonstrate. The platform reproduces the exploit within approved scope and returns the evidence with a verdict: requests, responses, and context. This is the "Find it. Fix it." side of VDR: continuous penetration testing, vulnerability report triage, and retesting after fixes.
  • Prove the risk. Exploitability validation and internet-reachability evidence make findings reportable. A component behind a firewall can still be internet-reachable when internet-originated input reaches it through an application or API, and the evidence captures that path.
  • Retest the fix. After the team ships a remediation, the platform replays the original attack. The retest result joins the finding's record, so the trail from finding to fix to verification stays connected.
  • Report in a structured form. Findings and test data are structured so they can be incorporated into a FedRAMP reporting workflow, not locked in a PDF.

Because testing is triggered by releases rather than by calendar, the evidence stays tied to recent releases. That currency is what makes the record auditable months later, provided each result records its scope, environment, date, and target release.

Where Casco for Government fits

Casco for Government is the federal offering of Casco, an AI-powered penetration-testing platform. It is FedRAMP Certified (20x, Class A), certified on November 5, 2026 per Casco's certification announcement, and Casco describes itself as the first agentic pentesting platform certified via FedRAMP 20x. It runs in an isolated AWS account, with model access through Amazon Bedrock in US-East, and every engineer on the team is a U.S. citizen.

For the VDR/VER migration specifically, Casco provides evidence for up to 42% of FedRAMP 20x controls, mapped to named requirements on the FedRAMP feature page: VDR-CSO-DET (detection), VDR-TFR-PVR (retesting after fixes), VER-EVA-ELX and VER-EVA-EIR (exploitability and internet reachability), and FRC-CSO-JSN (machine-readable evidence via FedRAMP JSON schemas). For Class A, that is 26 of 66 controls.

Two boundaries matter for honest planning. Casco provides evidence; the customer still defines scope, assigns the Potential Agency Impact N-rating (PAIN), ships fixes within required timelines, maintains the persistent and monthly reports, and completes the Security Decision Record and the independent assessment. Casco does not replace the 3PAO. And certification gives agencies a standardized, reusable security package; each agency makes its own authorization and purchasing decision. Casco is working from Class A toward Class C, the tier FedRAMP describes for moderate-impact use, but Class C is not achieved yet.

Operationally, the government offering covers web apps, APIs, cloud infrastructure, mobile applications, browser extensions, internal and external networks, and AI applications, agents, LLM workflows, and MCP servers. Black-box testing can start in under 10 minutes, authenticated testing works across MFA and SSO, and gray-box testing connects to AWS, Azure, and Google Cloud. Agency teams can request the certification package and capability statement through casco.com/gov.

Frequently Asked Questions

Does a passing retest prove a vulnerability is fully remediated? No. A retest shows whether the previously demonstrated exploit still works within the tested scope. That is strong, auditable evidence, but it is scoped evidence, not a universal guarantee about every variant or environment. Treat it as verification of the demonstrated attack path.

What exactly does Casco provide for FedRAMP 20x? Casco provides evidence for up to 42% of FedRAMP 20x controls, including named VDR/VER requirements such as VDR-CSO-DET, VER-EVA-ELX, and FRC-CSO-JSN. The customer completes the rest of the shared responsibility: scope definition, PAIN ratings, fix timelines, report maintenance, and the independent assessment.

Is Casco for Government FedRAMP authorized? Casco for Government is FedRAMP Certified (20x, Class A), not "authorized" in the Rev5 sense. Certification gives agencies a reusable security package. Each agency still makes its own authorization and purchasing decision.

How current is the evidence a team can show an auditor? Evidence is generated per approved release and per retest, so the record reflects the tested release rather than last year's snapshot. Record scope, environment, date, and target release with each result, and the audit trail holds up on review.

Conclusion

Auditable proof that exploitable findings were tested and fixed is no longer a nice-to-have for federal software teams. VDR/VER deadlines in December 2026 and March 2027 make continuous detection, exploitability validation, and machine-readable reporting part of maintaining FedRAMP certification itself. The teams that meet this well will be the ones whose testing produces the record automatically: a reproduced exploit, a documented fix, and a retest that replays the original attack, all tied to a named release. Casco for Government is built to work that way, and its FedRAMP 20x Class A certification means the evidence it produces is shaped for federal review. If your offering is facing the VDR/VER migration, start with the requirement mapping on casco.com/features/fedramp and see where your evidence gaps sit.

Related Articles