Apache SkyWalking MCP version 0.1.0 contains a critical Server-Side Request Forgery (SSRF) vulnerability and a GraphQL expression injection issue. These vulnerabilities allow an attacker to manipulate server requests and inject malicious GraphQL expressions. The issue is fixed in version 0.2.0, which users are advised to upgrade to.
CVE-2026-74842 is a server-side request forgery (SSRF) vulnerability in the Kira-Pgr PromptShopMCP product affecting the download_image function in the Image-Toolkit-MCP-Server component. Manipulating the image_url argument can cause the server to make unintended requests. The vulnerability can be exploited remotely and public exploit code is available. The product uses a rolling release model, so specific affected or fixed versions are not provided. The vendor has not yet responded to the issue report.
A flaw has been found in jkawamoto mcp-florence2 up to 0.3.13. Affected by this issue is the function get_images of the file src/mcp_florence2/__init__.py. This manipulation of the argument src causes server-side request forgery. The attack may be initiated remotely. The exploit has been published and may be used. It is recommended to change the configuration settings. The vendor explains: "For deployments where SSRF protection is required, I recommend routing all HTTP(S) requests through an SSRF-safe proxy server. This approach mitigates the vulnerability without requiring changes to the mcp-florence2 source code."
CVE-2026-19978 is an OS command injection vulnerability in the jiantao88 android-mcp-server component. The flaw exists in the child_process.exec function within build/index.js, where manipulation of certain input arguments can lead to command injection on the local host. Exploit code has been published. The product uses a rolling release model, so specific affected or patched versions are not available. A patch identified by commit 14e2bf27c88ba137e35cbb0c2a75f72b595bb98a is recommended to fix the issue. The CVSS 4.0 base score is 4.8, indicating medium severity.
A vulnerability was found in Jij-Inc Jij-MCP-Server 0.1.0. This affects the function PythonREPL.run of the file jij_mcp/python_repr.py of the component jm_check. The manipulation of the argument code results in code injection. It is possible to launch the attack remotely. The exploit has been made public and could be used. The project was informed of the problem early through an issue report but has not responded yet.
A vulnerability was identified in graphlit graphlit-mcp-server 1.0.1. This affects the function fetch of the file src/tools.ts of the component ssrf-test Endpoint. Such manipulation of the argument url leads to server-side request forgery. The attack may be launched remotely. The exploit is publicly available and might be used. The project was informed of the problem early through an issue report but has not responded yet.
CVE-2026-19956 is a server-side request forgery (SSRF) vulnerability found in gomarble-ai facebook-ads-mcp-server version 0.1.0. The vulnerability exists in the fetch_pagination_url function of the server.py file, allowing remote attackers to manipulate requests. A patch identified by commit 4e53875aa22e8991c2fa4a7660d86e1caba66659 is available to address this issue.
In the Linux kernel, the following vulnerability has been resolved: ublk: reset kernel-owned dev_info fields in ublk_ctrl_add_dev() ublk_ctrl_add_dev() memcpy()s the userspace ublksrv_ctrl_dev_info into ub->dev_info and then fixes up the fields the driver owns, but misses ->state and ->ublksrv_pid. A device added with ->state = UBLK_S_DEV_LIVE passes the "->state != UBLK_S_DEV_DEAD" test that ublk_stop_dev_unlocked() uses as its proxy for "a disk is attached", while ->ub_disk is still NULL, so DEL_DEV right after ADD_DEV oopses in del_gendisk(). UBLK_S_DEV_QUIESCED plus UBLK_F_USER_RECOVERY dies one step earlier, in ublk_force_abort_dev(). A poisoned ->state also gets START_USER_RECOVERY and the char device read/write path onto a device that was never started, and wedges START_DEV at -EEXIST. A poisoned ->ublksrv_pid just makes GET_DEV_INFO report an unrelated task as the ublk server. Reset both after the memcpy(), as ublk_detach_disk() does. Userspace only ever reads these back, so correcting them silently breaks nothing. ADD_DEV has copied ->state in unsanitized since ublk was merged, but back then it was harmless: the gendisk was allocated during ADD_DEV, and both teardown and the START_DEV -EEXIST check keyed off disk_live() rather than ->state. The oops became reachable once the disk allocation moved to START_DEV and those checks switched to ->state.
Token Optimizer MCP version prior to 5.1.0 contains an unauthenticated path traversal vulnerability in its dashboard session log API endpoints (/api/session-summary and /api/session-events). The endpoints accept a sessionId parameter that is directly concatenated into a filesystem path without proper sanitization, allowing attackers to read arbitrary .jsonl files on the server. This can expose sensitive session log data. The vulnerability has a CVSS score of 5.3 (medium severity) and is fixed in version 5.1.0.
Token Optimizer MCP version prior to 5.1.0 contains an OS command injection vulnerability in its smart_user tool. The vulnerability arises because the username argument in the get-user-info operation is interpolated directly into a shell command executed via execAsync(), allowing command substitution inside double quotes. This enables an attacker controlling the username parameter to execute arbitrary shell commands with the privileges of the MCP server user. The issue is identified as CWE-78 and has a CVSS 3.1 score of 8.4 (high severity). A fix is available in version 5.1.0.
CVE-2026-50027 is a critical authentication bypass vulnerability in doobidoo's mcp-memory-service versions prior to 10.67.1. The vulnerability allows unauthenticated remote attackers to perform unauthorized read, write, and delete operations on the memory store via HTTP routes under /api/documents/*. This occurs despite the server being configured with API key or OAuth authentication, due to inconsistent enforcement compared to the /api/memories endpoint. The issue is fixed in version 10.67.1.
CKAN MCP Server is a tool for querying CKAN open data portals. Prior to 0.4.112, canonicalizeParams in src/utils/cache.ts serializes request parameters with unescaped ampersand, equals-sign, and vertical-bar delimiters, allowing different logical parameter sets used by buildCacheKey to collide and an attacker to prime a shared cache with a response for a victim's distinct query. This issue is fixed in version 0.4.112.
CKAN MCP Server is a tool for querying CKAN open data portals. Prior to 0.4.112, the ckan_get_mqa_quality and ckan_get_mqa_quality_details tools in src/tools/quality.ts use isValidMqaServer to validate the server_url parameter with a prefix-only regular expression for dati.gov.it, allowing suffix-host and URL-userinfo values to target an attacker-controlled host and return a spoofed response. This issue is fixed in version 0.4.112.
CKAN MCP Server is a tool for querying CKAN open data portals. Prior to 0.4.112, error paths reflect raw upstream response bodies and internal exception messages back to the caller instead of a sanitized, generic message. When the server is pointed at (or redirected/SSRF'd to) a host that returns a non-CKAN response, or when an internal exception occurs, the caller receives verbatim upstream content and internal detail (hostnames, internal IPs, DB errors, stack fragments). This vulnerability is fixed in 0.4.112.
The Cortex MCP server (`neuro-cortex-memory`), a cross-platform persistent memory MCP, prior to version 3.17.1 treats the `CLAUDE_PROJECT_DIR` environment variable — automatically set by Claude Code to the currently open project directory — as a trusted Cortex developer checkout. When the `open_visualization` tool is invoked, `_find_dev_source()` resolves the user's active project directory as a candidate Cortex source root. The only validation performed by `_is_cortex_root()` is a check for the presence of an `mcp_server/` subdirectory and a `ui/unified-viz.html` file. An attacker who places these two marker files in a malicious repository can cause Cortex to execute an arbitrary `mcp_server/server/visualize_bootstrap.py` from that directory via `subprocess.run([sys.executable, ...])`, achieving code execution with the privileges of the victim's local user process. Version 3.17.1 fixes the issue.
CVE-2026-19770 is a server-side request forgery (SSRF) vulnerability in feedmob fm-mcp-servers version 0.0.3. The flaw exists in the downloadReport function of the Download Endpoint component, where manipulation of the downloadUrl argument allows SSRF attacks. Exploitation requires local access to the environment. The vulnerability has a medium severity score of 4.8 and public exploit code is available. The vendor has not yet responded or issued a patch.
A security flaw has been discovered in eyaushev swagger-testcase-mcp 5babb27c951fb404bc2b25ec80593616e49054e5. This vulnerability affects the function loadSource of the file src/utils/swagger-parser.ts of the component fetch_swagger. Performing a manipulation results in server-side request forgery. The attack is possible to be carried out remotely. The exploit has been released to the public and may be used for attacks. 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.
CVE-2026-19753 is a server-side request forgery (SSRF) vulnerability in Model Context Protocol mcp-rdf-explorer version 1.0.0. The flaw exists in the explore_url function within the MCP Server component, allowing remote attackers to manipulate the url argument to perform SSRF attacks. The vulnerability has a medium severity rating with a CVSS score of 6.9. Exploit code is publicly available, but no vendor response or patch has been provided as of the publication date.