CVE-2025-66025 Overview
Caido is a web security auditing toolkit used by penetration testers and security researchers. CVE-2025-66025 affects the Markdown renderer used in Caido's Findings page in versions prior to 0.53.0. The renderer improperly handles user-supplied Markdown and renders attacker-controlled links without confirmation. When a user opens a finding generated by the scanner or a plugin, clicking an injected link redirects the Caido application to an attacker-controlled domain. The flaw enables phishing-style attacks against Caido operators. The issue is patched in version 0.53.0.
Critical Impact
An attacker who can influence finding content can redirect Caido users to malicious domains through Markdown links, enabling credential phishing and follow-on social engineering.
Affected Products
- Caido web security auditing toolkit prior to version 0.53.0
- Findings page Markdown renderer component
- Any Caido plugin or scanner module that writes attacker-influenced content into findings
Discovery Timeline
- 2025-11-26 - CVE-2025-66025 published to NVD
- 2026-06-17 - Last updated in NVD database
Technical Details for CVE-2025-66025
Vulnerability Analysis
The vulnerability is an improper neutralization of special elements in output [CWE-74]. Caido's Findings page renders Markdown supplied by scanners and plugins. The renderer treats Markdown link syntax as trusted and produces clickable hyperlinks without confirming the destination with the operator.
An attacker who controls any input processed by a scanner or plugin can inject Markdown links pointing to arbitrary domains. When the finding is later reviewed, the operator sees a link that appears related to the audit target. Clicking the link navigates the Caido application to the attacker-controlled site.
Because Caido is used during offensive security engagements, operators routinely trust content surfaced by their own tooling. This trust boundary makes the injected link an effective phishing vector for credentials, session tokens, or additional payloads. The impact is limited to redirection and content disclosure to the attacker domain, without direct integrity or availability effects on the toolkit itself.
Root Cause
The root cause is missing link validation and missing user confirmation in the Markdown rendering pipeline on the Findings page. The renderer accepts any URL scheme and target embedded in Markdown syntax such as [label](https://attacker.example) and produces a live hyperlink without prompting the user.
Attack Vector
The attack requires network-adjacent input into a scanner or plugin and user interaction to click the rendered link. An attacker plants Markdown link syntax inside a response, header, or other artifact that a Caido scanner or plugin captures and writes into a finding. The finding is then presented in the UI. When the operator clicks the injected link, the Caido application follows the redirect to the attacker-controlled domain, where phishing content is served.
No verified public exploit code is available. The mechanism is described in the GitHub Security Advisory.
Detection Methods for CVE-2025-66025
Indicators of Compromise
- Outbound connections from the Caido host to domains unrelated to the current engagement scope
- Findings containing Markdown link syntax pointing to external domains not associated with the target application
- Browser or Caido navigation events immediately following the opening of a specific finding
Detection Strategies
- Review historical findings for Markdown links whose destination hostname does not match the audited target scope
- Correlate Caido process network telemetry with the timestamps of finding-open events to surface unexpected navigations
- Inspect scanner and plugin output for Markdown link patterns such as [text](url) before rendering
Monitoring Recommendations
- Log and alert on Caido client egress to newly observed domains during active engagements
- Track the Caido application version across analyst workstations and flag installs below 0.53.0
- Monitor plugin installations and updates for untrusted sources that could inject malicious Markdown
How to Mitigate CVE-2025-66025
Immediate Actions Required
- Upgrade all Caido installations to version 0.53.0 or later
- Audit installed plugins and remove any from untrusted publishers
- Instruct operators to treat links inside findings as untrusted and verify destinations before clicking
Patch Information
The issue is fixed in Caido 0.53.0. The patched Markdown renderer no longer follows attacker-controlled links without user confirmation. Details are published in the Caido GHSA-cf52-h5mw-gmc2 advisory.
Workarounds
- Avoid clicking hyperlinks rendered inside findings on Caido versions below 0.53.0
- Disable or limit scanners and plugins that write untrusted content into findings until the upgrade is applied
- Use host-level egress filtering to restrict Caido from reaching arbitrary external domains during engagements
Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.
