CVE-2026-28465: OpenClaw voice-call plugin webhook verification bypass in proxy configurations

CVE-2026-28465 · Severity: medium · CVSS 5.9 · Published 2026-03-05

OpenClaw's optional voice-call plugin handles incoming webhook calls for voice integration features. In certain proxy or forwarding setups, the plugin may incorrectly trust forwarded headers (like X-Forwarded-For) that are not validated by a trusted proxy, allowing an attacker to forge webhook requests and inject spoofed voice-call events into the system. This could disrupt voice-call functionality or inject malicious events into downstream processing.

Technologies: Openclaw, Openclaw Voice-Call. Vendors: npm, Openclaw.

Quick answers

Is CVE-2026-28465 exploited in the wild?
No. Junglewise Threat Intelligence has no report of CVE-2026-28465 being exploited in the wild.
How severe is CVE-2026-28465?
Medium severity, with a CVSS score of 5.9, published 5 March 2026.
Which products and versions are affected?
@clawdbot/voice-call (npm) SEMVER:introduced=0,last_affected=2026.1.24; @openclaw/voice-call (npm) SEMVER:introduced=0,fixed=2026.2.3; Openclaw Voice-Call < 2026.2.3.
Is there a fix for CVE-2026-28465?
Yes. @openclaw/voice-call (npm): upgrade to 2026.2.3 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
How is CVE-2026-28465 attacked?
It is attacked over the network, in a software library.
Am I affected by CVE-2026-28465?
Check whether you run @clawdbot/voice-call (npm), @openclaw/voice-call (npm), Openclaw or one of the others listed at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

How to fix

  • @openclaw/voice-call (npm): upgrade to 2026.2.3 or later
    npm install @openclaw/voice-call@2026.2.3

Technical details

The vulnerability stems from improper authentication of webhook requests in OpenClaw's voice-call plugin (CWE-287, CWE-290, CWE-345). The root cause is that some deployments implicitly trust forwarded headers (Forwarded / X-Forwarded-* headers) when determining request properties used during webhook signature verification. If those headers are not overwritten by a trusted proxy, a client can supply them directly to influence verification logic and bypass authentication checks. The attack requires network access to the webhook endpoint but no privileges or user interaction. Patched versions ignore forwarded headers by default unless explicitly allowlisted in configuration. The fix was released in @openclaw/voice-call >= 2026.2.3; the legacy package @clawdbot/voice-call has no patched version published and users must migrate. Workaround: strip Forwarded and X-Forwarded-* headers at the edge proxy.

Affected products

  • OpenClaw @openclaw/voice-call < 2026.2.3
  • OpenClaw @clawdbot/voice-call <= 2026.1.24

Identifiers

  • CVE-2026-28465
  • GHSA-3m3q-x3gj-f79x

Timeline

  • 2026-02-17: disclosed: Advisory published as GHSA-3m3q-x3gj-f79x
  • 2026-02-03: patched: Fix commit a749db9820eb6d6224032a5a34223d286d2dcc2f (hardening webhook verification)
  • 2026-02-17: other: CVE-2026-28465 assigned

References

This page as Markdown. Free to reuse under CC BY 4.0 with a link back.