{"vuln_id":"CVE-2026-133505","aliases":["GHSA-4cj4-hgwf-4hc4"],"description":"pydantic-ai: pydantic-ai-slim: pydantic-ai: Information disclosure via unsafe deserialization of tool results","published_at":"2026-11-20T07:34:58Z","modified_at":"2026-11-20T18:18:50.020000Z","first_seen_at":"2026-11-21T19:15:15.261390Z","severity":"medium","severity_source":"nvd","cvss_score":5.9,"cvss_vector":"CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N","cwe":["CWE-502","CWE-843","CWE-918"],"vendor":null,"product":null,"kev":{"is_kev":false,"date_added":null,"due_date":null,"ransomware":false},"epss":0.00702,"epss_percentile":0.54147,"nvd_status":"Received","enriched_at":null,"links":{"nvd":"https://nvd.nist.gov/vuln/detail/CVE-2026-133505"},"matched_alias":null,"items":[{"id":9673,"source":{"id":"nvd","name":"NVD CVE API 2.0"},"title":"CVE-2026-133505: When message history is serialized and loaded again, a tool result that contains a JSON object in the serialized shape of one of Pydantic AI's file types is re…","summary":"When message history is serialized and loaded again, a tool result that contains a JSON object in the serialized shape of one of Pydantic AI's file types is rebuilt as that file. A third party who controls data a tool returns, such as a record that an MCP server or a function tool returns from an external system, can use this to make the application download a URL on its local network, or make the model provider read an s3:// object or uploaded file with the application's credentials. Applications are affected when tool results can contain third-party data, message history containing those results is serialized and loaded again (stored history, history a UI adapter holds on the server, or Temporal, DBOS or Prefect durable execution), and the model is one Pydantic AI downloads files for, one that honors force_download, or one whose provider resolves non-HTTP file URLs or uploaded files. History that a client submits through the Vercel AI or AG-UI adapter in the default mode is not affected. The issue is fixed in pydantic-ai and pydantic-ai-slim 1.107.10 and 2.66.0.\n\nReferences:\nhttps://github.com/pydantic/pydantic-ai/security/advisories/GHSA-4cj4-hgwf-4hc4\nhttps://pydantic.dev/docs/ai/ui/overview/#trust-model-for-client-submitted-messages","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-133505","published_at":"2026-11-20T08:18:33.637000Z","updated_at":"2026-11-20T18:18:50.020000Z","fetched_at":"2026-11-21T19:15:31.485215Z","severity":"medium","cvss_score":5.9,"epss":0.00702,"is_kev":false,"has_exploit":false,"kev_due_date":null,"vendor":null,"product":null,"vuln_id":"CVE-2026-133505","vuln_ids":["CVE-2026-133505","GHSA-4CJ4-HGWF-4HC4"],"tags":["CWE-843","CWE-918"],"cluster_id":550,"links":{"self":"https://vulnwatch.ve1ve2.com/api/v1/items/9673","html":"https://vulnwatch.ve1ve2.com/item/9673"}},{"id":811,"source":{"id":"redhat","name":"Red Hat Security Data API"},"title":"CVE-2026-133505: pydantic-ai: pydantic-ai-slim: pydantic-ai: Information disclosure via unsafe deserialization of tool results","summary":"pydantic-ai: pydantic-ai-slim: pydantic-ai: Information disclosure via unsafe deserialization of tool results","url":"https://access.redhat.com/security/cve/cve-2026-133505","published_at":"2026-11-20T07:34:58Z","updated_at":null,"fetched_at":"2026-11-21T19:15:08.892859Z","severity":"medium","cvss_score":5.9,"epss":0.00702,"is_kev":false,"has_exploit":false,"kev_due_date":null,"vendor":"redhat","product":null,"vuln_id":"CVE-2026-133505","vuln_ids":["CVE-2026-133505","GHSA-4CJ4-HGWF-4HC4"],"tags":["redhat"],"cluster_id":550,"links":{"self":"https://vulnwatch.ve1ve2.com/api/v1/items/811","html":"https://vulnwatch.ve1ve2.com/item/811"}}]}