Loading story…
Tilt patched a high-severity flaw tracked as CVE-2026-55883 that allowed cross-site WebSocket hijacking of the local HUD server, exposing the /ws/view stream to reachable attackers when the interface was bound beyond localhost. Affected versions 0.24.0 through 0.37.3 could leak session state, Tiltfile contents, resource status, and live updates because the WebSocket relied on a CSRF token obtainable from the unauthenticated /api/websocket_token endpoint, and the upgrader accepted clients without an Origin header. The issue is rated CVSS 8.3 and is fixed in Tilt 0.37.4.
The remediation hardens Tilt's HTTP/HUD server beyond the WebSocket path by adding token-based authentication to sensitive and mutating API routes, validating Origin headers to block cross-origin abuse, setting the session cookie to SameSite=Strict, and applying the same token checks to proxy endpoints. Tilt also restricted /debug/pprof-style debug endpoints to loopback clients and updated CLI behavior to send the Tilt token in an X-Tilt-Token header, while warning that exposing the HUD on non-local interfaces has security implications. Users are advised to upgrade to 0.37.4, keep the HUD off the network where possible, and limit access if remote exposure is required.
Mallory correlates global threat intelligence with your attack surface — know if you’re exposed before adversaries strike.
2 events from the most recent confirmed update back to the earliest known activity.
CVE-2026-55883 was published describing a high-severity cross-site WebSocket hijacking issue in Tilt affecting versions 0.24.0 through 0.37.3. The advisory states the unauthenticated /api/websocket_token endpoint and missing Origin enforcement on /ws/view could let a network-reachable attacker access the full HUD view stream, and notes the issue is fixed in version 0.37.4.
A Tilt GitHub commit introduced token-based authentication for multiple HUD/API routes, Origin header validation, SameSite=Strict cookies, loopback-only access for debug endpoints, and matching token checks for proxy access. The change also updated the CLI to send the Tilt token in requests and documented that disabling HUD auth is only intended behind trusted reverse proxies or load balancers.
Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.
2 references tracked. Mallory keeps watching after this page renders.
Map indicators from this story to your assets and identify affected systems in minutes.
Every observed campaign, victim, and pivot linked to actors named in this story.
Malware, exploits, and IOCs connected to the activity described here.
YARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Get matching new stories delivered to your team as they break — not the next morning.
Ask questions about this story and take action on the answers.
Other stories Mallory is tracking that overlap on entities, sources, or topic labels.
Cline disclosed and fixed **CVE-2026-59723**, a high-severity cross-origin WebSocket hijacking flaw in the Hub dashboard server started by the `cline dashboard` command. In versions before **3.0.30**, the `/browser` WebSocket endpoint failed to properly validate the `Origin` header, and authorization could be bypassed when `ROOM_SECRET` was unset for local `127.0.0.1` binds. A malicious website could abuse the weakness to send `desktopCommand` frames to the local dashboard, exposing workspace state, changing MCP and provider settings, and potentially triggering command execution when a provider or model was configured. The fix landed through pull request **#11724** and commit `d09270940f5746f288cfc4a5039b46a2f4d5d01e`, which added host and origin allowlists, enforced authentication for WebSocket upgrades, unsafe HTTP methods, and non-public routes, and required a matching room secret when configured. The patch also normalized public URL handling for localhost and direct IP access, while tests were added to verify trusted local origins and reject untrusted hosts and origins. Cline advised users to upgrade to **3.0.30**, set `ROOM_SECRET` when the dashboard is not strictly local, validate origins, and limit provider and model configurations to reduce exposure.
Jul 9, 2026
Multiple high-severity vulnerabilities in **Open WebUI** allow attackers to plant malicious chat content that executes when viewed, leading to theft of JWTs and session tokens from `localStorage`, full account takeover, and compromise of sensitive AI assets. The flaws affect markdown rendering, citation and response iframe handling, and HTML file-serving paths, including cases where attacker-controlled content was rendered with dangerous HTML processing or iframe sandbox settings such as `allow-scripts allow-forms allow-same-origin`. Reported impacts include exposure of DOM data, `indexedDB`, cookies, backend API keys, model parameter files, private databases, and other enterprise AI resources. Several reports warn that if an administrator opens a malicious shared transcript or chat, the attack can escalate from stored XSS to **remote code execution** through Open WebUI's administrative features and Python-based functions or plugins, potentially yielding code execution on the Docker container or local host. The issues were addressed across releases including **v0.6.6** and **v0.7.0** by tightening file ownership checks, removing or making optional same-origin iframe access, and recommending stronger HTML sanitization and restrictive CSP settings; one report also provided a database query to hunt for stored chats containing suspicious `data:text/html` or `javascript:` embed payloads.
Jul 7, 2026
Hoppscotch, the open source API development platform, patched two high-severity vulnerabilities in version **2026.3.0**. **CVE-2026-34931** is an improper loopback `redirect_uri` validation flaw in the device-login flow that creates an open redirect condition, allowing token exfiltration and potentially letting an attacker sign in as a victim and take over the victim’s account. The issue is classified as **CWE-601** and is remotely exploitable with low attack complexity and no privileges required. The same release also fixed **CVE-2026-34932**, a stored cross-site scripting vulnerability in mock server responses on the backend origin. The flaw, classified as **CWE-79**, can be used to trigger cross-site request forgery and requires user interaction, but is likewise network exploitable with low attack complexity and no prior privileges. Both issues affect Hoppscotch versions prior to **2026.3.0**, making upgrade to the patched release the key remediation step.
Jun 29, 2026