---
id: "CVE-2026-119712"
kind: "vulnerability"
aliases: ["CVE-2026-119712"]
title: "PipesHub: indexing service fetches internal URLs embedded in documents"
severity: "medium"
cvss: 6.4
epss: null
exploited: false
status: "open"
category: "ai-tooling"
products: []
ecosystems: []
tags: ["ai-indexing", "ai-tooling", "cloud-metadata", "pipeshub", "self-hosted"]
not_affected_if: ["no-internal-network-reach", "version-0.10.0-or-later"]
published_at: "2026-10-30T16:17:54+00:00"
first_seen_at: "2026-10-30T16:31:26+00:00"
updated_at: "2026-10-30T16:32:21+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-119712", "first_seen_at": "2026-10-30T16:31:26+00:00"}]
---

# PipesHub: indexing service fetches internal URLs embedded in documents

PipesHub before 0.10.0 has a server-side request forgery flaw in its image parser, which fetches any image URL embedded in an uploaded document, synced mailbox, or web connector page. An attacker can make the indexing service reach loopback, private-network, or cloud metadata endpoints and get the responses embedded into indexed content.

_Summary, check and fix written by a model from the sources; labelled as such._

## How to check

- Check the installed PipesHub version; it is affected before `0.10.0`.
- Look for the `ImageParser._fetch_single_url` code path in the deployed indexing service.
- Confirm whether the indexing service can reach internal addresses such as loopback, private ranges, or cloud metadata endpoints.

## Fix

- PipesHub: move to `0.10.0` or later.

## Timeline

- 2026-10-30T16:31:26+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-119712

## From the sources

PipesHub before 0.10.0 contains a server-side request forgery vulnerability in ImageParser._fetch_single_url that allows attackers to make the indexing service fetch internal addresses by embedding image URLs in documents. Attackers can supply Markdown or HTML via uploads, synced mailboxes, or web connectors to reach loopback, private-network, or cloud metadata endpoints, with image responses embedded into indexed content.

---

_Free to read at a human pace. An API with keys and an MCP server are in preparation; early access: hello@infrastack.ai_