# vLLM: SSRF + arbitrary local file read in MiMoV2OmniMultiModalProcessor \`\_fetch_image\` and audio loader bypass MediaConnector protections

Advisory record · `GHSA-4hhp-h66f-j5j7`

- **Identifiers:** `GHSA-4hhp-h66f-j5j7` · `CVE-2026-73560` · `PYSEC-2026-3934`
- **State:** Confirmed
- **Severity:** CVSS 6.5 Medium
- **Critical:** No
- **Weakness (CWE):** [CWE-918](https://cwe.mitre.org/data/definitions/918.html)
- **Tags:** [Server-side request forgery (SSRF)](https://agentprime.io/advisories/tag/ssrf.md)
- **Published:** 8 Sept 2026
- **Updated:** 10 Sept 2026
- **Reviewed by a person:** No

Published by the GitHub Advisory Database at 8 Sept 2026, 20:42 UTC. agentprime has not analysed this yet, so whether it affects any app is not known.

## Affected packages

- `vllm` (PyPI): affected `<0.26.0`; fixed in `0.26.0`

## Machine-written summary

Summary written by Claude Haiku from the source advisory. Not reviewed by a person.

> GHSA-4hhp-h66f-j5j7: vllm versions before 0.26.0 are affected. The MiMoV2Omni multimodal processor fetches user-supplied image and audio URLs directly and opens local paths without the SSRF and local-path checks that the connector applies. An attacker who can pass a string into multi_modal_data (for example through library use of LLM.generate) could make the server request internal or cloud-metadata addresses or read local files the vLLM process can access. Upgrade to vllm 0.26.0 or later.

Written 1 Nov 2026.

## Not analysed yet

We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.

## Severity and exploitation

- 6.5 Medium, CVSS 3.1: `CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N`, from CNA (GitHub_M)

- **CISA KEV:** Not listed.
- **EPSS:** 0.0041, the 35th percentile, on 9 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

## Timeline

- 31 Oct 2026, 14:10 UTC: First seen by agentprime
- 31 Oct 2026, 14:10 UTC: [Confirmed by OSV](https://osv.dev/vulnerability/GHSA-4hhp-h66f-j5j7)

## Sources

- [GitHub Advisory Database: GHSA-4hhp-h66f-j5j7](https://github.com/advisories/GHSA-4hhp-h66f-j5j7)
- [OSV: GHSA-4hhp-h66f-j5j7](https://osv.dev/vulnerability/GHSA-4hhp-h66f-j5j7)
- <https://github.com/vllm-project/vllm/security/advisories/GHSA-4hhp-h66f-j5j7>
- <https://nvd.nist.gov/vuln/detail/CVE-2026-73560>
- <https://github.com/vllm-project/vllm/pull/43117>
- <https://github.com/vllm-project/vllm/commit/54503ecec0f3ac31e5ecfc5f28652e4cc42307b5>
- <https://github.com/vllm-project/vllm>
- <https://github.com/vllm-project/vllm/releases/tag/v0.26.0>

Facts on this page come from the GitHub Advisory Database (CC-BY 4.0), OSV and the other sources listed, as published. We don’t reproduce full advisory descriptions, which can contain exploit code.

## Check whether this affects you

A Match means an installed version is in the affected range above. Most don’t affect the app; agentprime hasn’t analysed this one yet.

- With the agentprime MCP server (<https://api.agentprime.io/mcp>), ask your agent to check the project: `check_lockfile` sends a lockfile, `check_packages` a list of installed packages, and both return the Matches and the upgrades that clear them. `get_advisory` returns this record. Setup: <https://agentprime.io/mcp.md>
- In a browser: <https://agentprime.io/check/> reads your lockfile in the browser, never uploading it.
- Whether it touches your code, with file and line: the Claude Code plugin, <https://agentprime.io/plugin.md>
- As JSON: <https://api.agentprime.io/v1/advisories/GHSA-4hhp-h66f-j5j7>, or `POST https://api.agentprime.io/v1/match` with your packages (<https://agentprime.io/api.md>)

---

The web page: <https://agentprime.io/advisories/GHSA-4hhp-h66f-j5j7/>. Everything for agents: <https://agentprime.io/llms.txt>.