/

CVE-2026-87821

Severity:

HIGH RISK

(7.1)

SSRF (CWE-918) in Lara Dashboard Markdown Fetch Endpoint

CVE-2026-87821

Deferred

Vulnerability type

Category:

Server-Side Request Forgery

CWE:

CWE-918

Impact:

Unauthorized access to internal services and sensitive network resources

Severity:

HIGH RISK

(7.1)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

Lara Dashboard contains a server-side request forgery vulnerability in the authenticated Markdown fetch endpoint, allowing users to make the server request attacker-controlled URLs and read response bodies. This may expose internal services, cloud metadata, and sensitive credentials when outbound requests are not restricted by destination or redirect policy.

Affected context

The vulnerability occurs in Lara Dashboard deployments that expose the authenticated

POST /api/admin/builder/markdown/fetch endpoint and allow user-supplied URLs to be

fetched by the server. Risk increases when outbound requests can reach private network

addresses, cloud metadata services, or redirected destinations without host validation.

Exploit preconditions

  • Attack surface: The authenticated Markdown fetch API endpoint
  • Required access: Any account permitted to invoke the endpoint
  • Trigger point: Submission of a URL for server-side Markdown retrieval
  • Why exploitation is possible: The server makes outbound requests using attacker-controlled

destinations and returns response content without sufficient host and redirect restrictions

Why this happens

The underlying bug is an unchecked server-side HTTP request based on user-controlled URL

input. The endpoint does not adequately constrain permitted hosts or prevent redirects from

reaching restricted destinations, allowing the application server to act as a proxy into

networks that may not be directly reachable by the requester. This behavior maps to

CWE-918: Server-Side Request Forgery, where external input influences requests made

from a trusted server context.

Potential impact

  • Attackers may read responses from internal HTTP services
  • Cloud instance metadata and IAM credentials could be exposed
  • Internal administrative interfaces may become reachable
  • Network topology and service information may be disclosed
  • Compromised credentials could enable further unauthorized access

General mitigation guidance

Mitigation generally focuses on reducing trust in user-supplied URLs and enforcing

destination-aware outbound request policies.

Common defensive patterns include:

  • Allowing requests only to an explicit, minimal set of approved hosts and schemes
  • Resolving and validating destinations before connecting, including protection against DNS rebinding
  • Blocking loopback, link-local, private, reserved, and other restricted address ranges
  • Revalidating every redirect and limiting redirect behavior by default
  • Avoiding direct access to cloud metadata services where platform controls support that separation
  • Applying network egress controls and least-privilege service identities
  • Returning only the minimum content required by the feature rather than arbitrary response bodies

How teams detect issues like this

Organizations can look for SSRF risks through:

  • Static analysis rules identifying HTTP clients whose destinations derive from request parameters
  • Code review checks for missing scheme, host, IP-range, DNS-resolution, and redirect validation
  • Runtime monitoring of outbound requests to private, loopback, link-local, and metadata addresses
  • Network policy enforcement that restricts application egress to approved destinations
  • Alerts for unusual request volume, unexpected internal hosts, or metadata-service access attempts
  • Automated security review tools that flag user-controlled URLs passed to server-side request libraries

Frequently asked

questions

An attacker generally needs an authenticated account that can invoke the Markdown fetch

endpoint. The required privilege level depends on how access to that feature is assigned.

It may, when the application server has network access to those destinations and outbound

filtering does not block them. Redirects can also expand the request path if they are not

independently validated.

Use an allowlist-based outbound request design, validate resolved destinations, block restricted

address ranges, and apply the same checks to every redirect.

Treat all user-provided URLs as untrusted, enforce egress policies at both application and

network layers, limit service identity privileges, and monitor outbound requests for anomalous

internal-resource access.

Find out what's already
exploitable in your codebase.