CVE-2026-19752 is a server-side request forgery (SSRF) vulnerability in the EnzoVezzaro mcp-dominican-layer project affecting the parse-pdf function in src/index.ts. The vulnerability allows remote attackers to manipulate the pdfUrl argument to induce the server to make unauthorized requests. The issue affects versions up to commit 39dd373786712650097ad31db27d5c477c8f9c82. The vulnerability has a medium severity rating with a CVSS 4.0 score of 5.3. Although an exploit has been publicly disclosed, the project has not yet responded or issued a patch.
A flaw has been found in EnzoVezzaro mcp-dominican-layer up to 39dd373786712650097ad31db27d5c477c8f9c82. The affected element is the function axios.get of the file src/index.ts of the component parse-csv tool. This manipulation of the argument csvUrl causes server-side request forgery. The attack is possible to be carried out remotely. The exploit has been published and may be used. This product adopts a rolling release strategy to maintain continuous delivery. Therefore, version details for affected or updated releases cannot be specified. The project was informed of the problem early through an issue report but has not responded yet.
The MCP Python SDK's experimental task handlers feature, when enabled, does not verify session ownership of tasks. This allows any connected client to list, read results from, and cancel tasks created by other clients on the same server. The vulnerability affects servers that enable the experimental tasks feature and use default handlers without session validation. It is fixed in version 1.27.2 and later, which embed session markers in task IDs and restrict task access to the owning session.
The MCP Python SDK's deprecated WebSocket server transport does not validate Host or Origin headers during the WebSocket handshake, allowing any web page to connect and issue JSON-RPC requests if the server is exposed without additional protections. This affects applications that explicitly use the deprecated websocket_server transport, which is not part of the MCP specification and must be manually integrated. The vulnerability can lead to unauthorized enumeration and invocation of server tools and resource access depending on the server's exposure. The issue is fixed in version 1.28.1 by adding optional Host and Origin validation, but upgrading alone does not enable validation unless configured. The recommended mitigation is to migrate away from the deprecated transport to Streamable HTTP, which enforces these protections automatically. The WebSocket transport is removed entirely in version 2.
Repomix MCP server's attach_packed_output tool can bypass secret scanning when reading local .json, .txt, .md, or .xml files. This allows reading arbitrary local files without the secret check enforced by the file_system_read_file tool. The issue affects repomix versions from 0.2.1 up to but not including 1.14.1 and is fixed in version 1.14.1.
A vulnerability in openclaw-cli versions up to 2026.4.21 allows a non-owner loopback client to spoof owner context metadata via request headers. This could enable unauthorized access to owner-gated operations. The issue is fixed in version 2026.4.22 by issuing separate bearer tokens for owner and non-owner contexts and no longer trusting the spoofable sender-owner header.
OpenClaw CLI is vulnerable to local file exfiltration via injection of MEDIA: directives in MCP tool result content. A malicious or compromised MCP tool server can specify arbitrary file paths within allowed directories such as the OS temporary directory, causing OpenClaw to read and send those files to external messaging channels like Discord, Slack, Telegram, and WhatsApp without user awareness. This vulnerability affects all OpenClaw versions up to and including 2026.7.1. The flaw arises from unvalidated extraction of MEDIA: tokens, an overly broad default allowlist including os.tmpdir(), and delivery of file contents to external channels. Secondary attack vectors include fallback on image content paths and support for file:// URI schemes. The impact includes theft of sensitive temporary files and silent exfiltration to attacker-controlled channels. A patch is available to remediate this issue.
A critical vulnerability (CVE-2026-32871) exists in Omnara's FastMCP OpenAPI Provider where path parameters are substituted into URL templates without URL encoding. This allows an attacker to inject path traversal sequences (e.g., '../') that are resolved by urljoin(), enabling authenticated Server-Side Request Forgery (SSRF) to access arbitrary backend endpoints with the MCP provider's authorization context.
CVE-2026-44694 is an authenticated Server-Side Request Forgery (SSRF) vulnerability in n8n-mcp affecting webhook trigger tools, the n8n API client, and URLs supplied via the x-n8n-url header in multi-tenant HTTP mode. An attacker with MCP session access can make HTTP requests from the n8n-mcp host to internal services and cloud metadata endpoints, potentially exfiltrating sensitive credentials. The vulnerability affects multi-tenant and single-tenant deployments, as well as stdio deployments via prompt injection. The issue is fixed in n8n-mcp version 2.50.2.
A Server-Side Request Forgery (SSRF) vulnerability exists in the n8n-mcp SDK embedder path versions 2.47.4 through 2.47.13. The synchronous URL validator in SSRFProtection.validateUrlSync() fails to check IPv4-mapped IPv6 addresses, allowing bypass of protections against cloud metadata, localhost, and private IP ranges. An attacker controlling the n8nApiUrl parameter can cause the server to make HTTP requests to internal or cloud metadata endpoints and receive response data, with the n8nApiKey header forwarded to the attacker-controlled target. This vulnerability is fixed in version 2.47.14.
n8n-mcp versions 2.14.1 up to but not including 2.47.13 running in HTTP transport mode log sensitive tool-call arguments including credential material such as bearer tokens, OAuth credentials, API keys, and webhook authentication headers. This logging occurs before any redaction, potentially exposing secrets in server logs accessible beyond the request trust boundary. The vulnerability requires authentication and is mitigated in stdio transport mode. A fix is available in version 2.47.13 that prevents logging sensitive values by summarizing arguments without recording raw data.
A vulnerability in n8n-mcp versions 2.14.1 through 2.47.10 allows sensitive request data to be logged on unauthorized POST /mcp HTTP requests. This includes bearer tokens, per-tenant API keys, and JSON-RPC payloads. Although unauthorized requests are rejected, sensitive information is still recorded in server logs, potentially exposing it if logs are accessible outside the trusted boundary. The issue affects deployments using HTTP transport mode and is fixed in version 2.47.11.
An authenticated Server-Side Request Forgery (SSRF) vulnerability exists in n8n-mcp versions up to 2.47.3. An attacker with a valid AUTH_TOKEN can cause the server to make HTTP requests to arbitrary URLs via multi-tenant HTTP headers, potentially accessing sensitive internal resources such as cloud metadata endpoints and internal network services. The issue affects multi-tenant HTTP deployments and is fixed in version 2.47.4.
### Summary
Several HTTP transport endpoints in n8n-mcp lacked proper authentication, and the health check endpoint exposed sensitive operational metadata without credentials.
### Impact
An unauthenticated attacker with network access to the n8n-mcp HTTP server could disrupt active MCP sessions and gather information useful for further attacks.
### Patches
Fixed in **v2.47.6**. All MCP session endpoints now require Bearer authentication. The health check endpoint has been reduced to a minimal liveness response.
### Workarounds
If you cannot upgrade immediately:
- **Restrict network access** to the HTTP server using firewall rules, reverse proxy IP allowlists, or a VPN so that only trusted clients can reach it.
- **Use stdio mode** (`MCP_MODE=stdio`) instead of HTTP mode. The stdio transport does not expose any HTTP endpoints and is unaffected by this vulnerability.
Upgrading to v2.47.6 is still strongly recommended.
### Credit
Reported by @yotampe-pluto.
## Impact
`n8n-mcp` versions before 2.50.1 contained three independently-reported issues affecting deployments that run the n8n API integration:
1. **Caller-supplied identifiers were not validated before being used as URL path segments** by the n8n API client. An authenticated MCP caller passing a crafted workflow id could cause outbound requests carrying the configured n8n API key to land on other same-origin endpoints, bypassing handler-level access controls (including `DISABLED_TOOLS`).
2. **Validated webhook, form, and chat trigger URLs followed redirects.** A URL that passed initial validation could redirect the outbound request to a host that would otherwise have been rejected, with the response body returned to the caller. Reachable as non-blind SSRF over authenticated MCP calls.
3. **Mutation telemetry stored unredacted operation payloads.** On instances running with the default opt-in telemetry, partial-update operation diffs were uploaded without redaction. Operation values can carry the same node-parameter values the workflow contains, including bearer tokens, API keys, and webhook secrets.
## Severity
CVSS 8.3 (HIGH). Exploitation requires an authenticated MCP caller and an n8n API integration configured with an n8n API key.
## Patched versions
Upgrade to `n8n-mcp >= 2.50.1`.
## Workarounds
- For issues (1) and (2): restrict network access to the HTTP transport (firewall, reverse-proxy ACL, or VPN) so only trusted callers can reach the MCP HTTP port; or switch to stdio mode, which exposes no HTTP surface for these issues.
- For issue (3): set `N8N_MCP_TELEMETRY_DISABLED=true` in the environment before starting the server, or run `npx n8n-mcp telemetry disable` once.
## Credit
Reported by @cybercraftsolutionsllc.
CVE-2026-44681 is an open redirect vulnerability in Authlib's OpenIDImplicitGrant and OpenIDHybridGrant authorization endpoints. An attacker can cause the authorization server to issue an HTTP 302 redirect to an arbitrary URL by submitting an authorization request that omits the required 'openid' scope. This occurs because the scope validation error redirects to an unvalidated redirect_uri supplied by the attacker. The flaw affects Authlib versions from 2.1.0 up to but not including 2.14.0_4 and is fixed in version 2.14.0_4. The vulnerability enables phishing attacks leveraging the trusted authorization server domain but does not leak tokens or codes directly.
FastMCP's authentication integration with Entra ID allows the MCP server to act both as an OAuth client and authorization server. Due to Entra ID's lack of Dynamic Client Registration support, the FastMCP server uses a static app registration for authorization. After a user consents to this static client, Entra ID sets a persistent authorization cookie. An attacker controlling their own MCP client can exploit this by initiating authorization flows that reuse the victim's consented static client ID, potentially leading to confused deputy account takeover.