CVE-2026-25904
Official description Straight from the sourceThe vendor's or NVD's own wording, published unedited. Authoritative, but often terse — it says what broke, rarely what to do.
NVD · uneditedThe Pydantic-AI MCP Run Python tool configures the Deno sandbox with an overly permissive configuration that allows the underlying Python code to access the localhost interface of the host to perform SSRF attacks. Note - the "mcp-run-python" project is archived and unlikely to receive a fix.
Technical summary Written by usOur analysis, written from the advisory, the CVSS vector and the affected-version data. It adds context the advisory leaves out, and never invents facts that are not in the source.
dbcve analysis · moderate confidenceThe Pydantic-AI MCP Run Python tool misconfigures the Deno sandbox with overly permissive settings that permit the executed Python code to make network requests to the localhost interface of the host system. This enables Server-Side Request Forgery (SSRF) attacks, allowing an attacker to access internal services, APIs, or infrastructure reachable on the host's loopback interface.
Verify against the referenced sources before acting — the references below are authoritative for this CVE, this summary is not.
CVSS breakdown How the score is builtThe industry scoring standard. It rates how the flaw is reached, what it takes to exploit, and what an attacker gains — the score is derived from those, not the other way round.
From the vector- Attack vector
- Network
- Complexity
- High
- Privileges
- None
- User interaction
- Required
- Scope
- Changed
- Confidentiality
- Low
- Integrity
- Low
- Availability
- Low
CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:C/C:L/I:L/A:L
Am I affected? How to checkSteps we derive from the advisory and the affected-version data, so you can decide whether this CVE reaches your setup. They are a guide, not a scan — your own configuration is the authority.
dbcve checksWork through these to decide whether this CVE applies to you.
-
Identify if mcp-run-python tool is in useSearch for mcp-run-python in installed packages, project dependencies, or running processes. Check for any Python execution paths that route through the Pydantic-AI MCP server.Affected if The mcp-run-python tool or Pydantic-AI MCP server with Python execution capability is installed and active in the environment.
-
Verify if network requests to localhost are permittedExecute a test Python script via the tool that attempts to make an HTTP request to http://127.0.0.1 or http://localhost on common internal ports (e.g., 8080, 3000, 5432).Affected if The executed Python code can successfully connect to localhost addresses, indicating the sandbox permits loopback network access.
-
Check for presence of Deno sandbox configurationLocate any configuration files related to the MCP server or Deno runtime. Look for sandbox settings, permissions flags, or network allowlists.Affected if Deno sandbox permissions are configured to allow network access to the loopback interface (127.0.0.1 or ::1).
-
Confirm the execution environment scopeDetermine whether the Python code execution runs locally on the same host as the MCP server or in an isolated environment (container, VM). Test if the tool can reach local services running on the host.Affected if The execution environment can access services bound to localhost on the host system, enabling potential internal service enumeration.
If the mcp-run-python tool is active and network requests to 127.0.0.1/localhost succeed from executed Python code, the environment is vulnerable to the SSRF misconfiguration described in this CVE.
Generated from the published advisory. Verify against your own configuration.
Remediation Closing itWhat it takes to close this. Where a vendor fix exists we point at it; where none exists we say so plainly, and can build one. Effort estimates are scoped from the advisory, not from your codebase.
From vendor dataSince the project is archived and will not receive a fix, discontinue use of the mcp-run-python tool or implement network isolation (firewall rules, container network policies) to block outbound connections from the execution environment to internal/localhost addresses.
- Consultation2.0 h
- Implementation4.0 h
- Testing3.0 h
- Review / QA1.0 h
An estimate, not a bill — we confirm scope with you before any work starts. Need it this week? Rush from $2,800.
Scan for this in your stack
Free · runs locallyCheck whether your project pulls in CVE-2026-25904 — or any other known-vulnerable package — straight from your lock files. Free and open source; it runs locally and uploads nothing.
References Go to the primary sourcePrimary sources — vendor advisories, patches and trackers. Where our summary and a reference disagree, the reference wins.
Primary sourcesPractitioner notes
ContributedPeer-ranked notes from engineers who’ve handled CVE-2026-25904 in production — separate from our analysis above.
The advisory tells you what broke. It rarely tells you what actually worked. If you’ve dealt with this one, that detail is what the next engineer is searching for.
- The version that genuinely resolved it — not the one the vendor claimed
- A config change or rule that shut the vector down
- A gotcha in the upgrade path that cost you an afternoon
No notes yet
Be the first to add a field note for this CVE — a mitigation you’ve verified, a version caveat, or a link to a working fix. Sign in above to contribute.
A place for practitioners to share what actually worked: a mitigation you’ve tested, a configuration change, a version- or environment-specific caveat, or a link to a verified patch. The most useful notes rise to the top as peers upvote them, so the signal stays high.
- Verified mitigations, workarounds, and config changes
- Version or environment caveats, and links to real fixes
- No weaponised exploit code, or anything meant to cause harm
- No spam, self-promotion, credentials, or personal data