CVE-2026-133505: Pydantic AI tool results become file downloads on history reload
- Source
- NVD(opens in a new tab)
- Identifier
- CVE-2026-133505
- Published
- Record updated
- Known exploitation
- Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
Summary
CVE-2026-133505 affects Pydantic AI. When message history is serialized and loaded again, a tool result containing a JSON object in the serialized shape of one of Pydantic AI's file types is rebuilt as that file. A third party controlling data returned by a tool, such as from an MCP server or function tool, can make the application download a URL on its local network, or make the model provider read an s3:// object or uploaded file using the application's credentials. Applications are affected when tool results can contain third-party data and the affected history is serialized and loaded again, including through Temporal, DBOS or Prefect durable execution; history submitted through the Vercel AI or AG-UI adapter in default mode is not affected.
Mitigation
The issue is fixed in pydantic-ai and pydantic-ai-slim 1.107.10 and 2.66.0.
Topics
Related items
- MediumCVE-2026-54026: LibreChat message injection via conversationId in request bodySimilar attack · NVD/CVE Database
- MediumCVE-2026-54046: LibreChat unauthenticated account enumeration via /api/user/verifySimilar attack · NVD/CVE Database
- HighCVE-2026-54044: LibreChat credential exposure through agent action domain repointingSimilar attack · NVD/CVE Database
- MediumCVE-2026-54043: LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.7…Similar attack · NVD/CVE Database
- MediumCVE-2026-54042: LibreChat 2FA verification unlimited code guessing via temp token endpointSimilar attack · NVD/CVE Database