---
id: CVE-2026-45019
title: >-
  Chainlit is a Python framework for building production-ready conversational AI
  applications
summary: >-
  Chainlit is a Python framework for building production-ready conversational AI
  applications. From 2.4.0rc0 until 2.12.0, Chainlit deployments with
  features.mcp.enabled set to true in .chainlit/config.toml expose the POST /mcp
  endpoint wi…
severity: high
cvss: 7.2
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:L/I:L/A:N'
cwe:
  - CWE-918
vendor: chainlit
product: chainlit
affected:
  - 'chainlit >= 2.4.0rc0, <= 2.11.1'
patched:
  - chainlit 2.12.0
published: '2026-08-25'
updated: '2026-09-09'
sourceUpdated: '2026-09-09T21:07:31.353'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-45019'
references:
  - url: >-
      https://github.com/Chainlit/chainlit/blob/2.12.0/docs/security-advisory-2026-mcp.md#spl-2026-002--ssrf-via-mcp-streamable-http--sse
    label: security-advisories@github.com
  - url: >-
      https://github.com/Chainlit/chainlit/commit/0565fd0eccb915fce159929598b053ed79f6e0c9
    label: security-advisories@github.com
  - url: 'https://github.com/Chainlit/chainlit/releases/tag/2.12.0'
    label: security-advisories@github.com
  - url: >-
      https://github.com/Chainlit/chainlit/security/advisories/GHSA-hvfh-5mj3-5f3j
    label: security-advisories@github.com
  - url: 'https://github.com/advisories/GHSA-hvfh-5mj3-5f3j'
  - url: 'https://github.com/Chainlit/chainlit'
  - url: 'https://pypi.org/project/chainlit'
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-45019'
tags:
  - nvd
  - ghsa
  - pip
  - osv
epss: 0.00434
epssPercentile: 0.35022
aliases:
  - GHSA-hvfh-5mj3-5f3j
  - PYSEC-2026-3811
ecosystem: pip
ingestedAt: '2026-08-25T19:31:03.091Z'
---

## Overview

Chainlit is a Python framework for building production-ready conversational AI applications. From 2.4.0rc0 until 2.12.0, Chainlit deployments with features.mcp.enabled set to true in .chainlit/config.toml expose the POST /mcp endpoint without requiring authentication. For sse and streamable-http transports, ConnectSseMCPRequest and ConnectStreamableHttpMCPRequest in backend/chainlit/types.py accept a user-controlled url and optional headers dictionary without scheme validation, private-address filtering, or an allowlist. The connect_mcp handler in backend/chainlit/server.py passes these values to sse_client() or streamablehttp_client(), allowing the Chainlit server to make blind outbound requests to arbitrary internal or external services, including cloud metadata endpoints, with attacker-controlled Authorization and Cookie headers. The SSE URL sink has existed since 2.4.0rc0, while attacker-controlled header forwarding and streamable-http support were added in 2.6.4. The response is consumed internally and not returned, but the attacker can issue state-changing authenticated requests, discover internal services, scan ports, and probe metadata endpoints. This issue is fixed in version 2.12.0.

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.

## Package advisory (CVE-2026-45019)

Affected packages:

- `chainlit >= 2.4.0rc0, <= 2.11.1`

Patched in:

- `chainlit 2.12.0`

Source: https://github.com/advisories/GHSA-hvfh-5mj3-5f3j