Retry prompts leak into Pydantic AI OpenTelemetry telemetry
Key facts
- Attack vector
- Network
- Privileges required
- Low
- User interaction
- None
- Attack complexity
- Low
- Attack requirements
- Present
- Impact type
- Information disclosure
- Confidentiality · integrity · availability
- Low · None · None
- Fix status
- Partial Red Hat: 2 without a fix
Attack profile from CVSS 4.0 · CNA. Fix status from CNA, GitHub and Red Hat. Impact type from CWE-532.
Summary
The pydantic-ai and pydantic-ai-slim PyPI packages from 0.3.4 before 1.107.4 and from 2.0.0b1 before 2.27.1 are affected. With include_content=False, OpenTelemetry instrumentation still exports retry prompts outside tool calls without redaction (CWE-532). Agents using NativeOutput, PromptedOutput or output validators can expose validation feedback, including invalid model values, to readers of the telemetry backend.
Written with Claude Haiku 5.5 from the sources below · 2026-10-29
Description from the CNA
Pydantic AI is a Python agent framework for building applications and workflows with Generative AI. From 0.3.4 until 1.107.4 and 2.27.1, OpenTelemetry instrumentation configured with InstrumentationSettings(include_content=False) can export retry prompts outside tool calls in gen_ai.input.messages and pydantic_ai.all_messages. Agents using NativeOutput, PromptedOutput, or output validators on text output can therefore disclose validation feedback, including invalid model values quoted by that feedback, to readers of the telemetry backend. Tool-call retries and deployments that do not use include_content=False are not affected by this specific path. This issue is fixed in versions 1.107.4 and 2.27.1.
Why it matters now
SSVC decision points
CISA ADP · SSVC 2.0.3
- Exploitation
- None
- Automatable
- No
- Technical impact
- Partial
The decision (Track, Attend, Act) also depends on your mission and well-being impact; see CISA's SSVC guide.
Affected products
| Vendor | Product | Affected versions | Fixed in | Status | Source |
|---|---|---|---|---|---|
| pydantic | pydantic-ai | 0.3.4 to before 1.107.4; 2.0.0b1 to before 2.27.1 | 1.107.4, 2.27.1 | Patch available | CNA |
| pydantic | pydantic-ai-slim | 0.3.4 to before 1.107.4; 2.0.0b1 to before 2.27.1 | 1.107.4, 2.27.1 | Patch available | CNA |
| — | pydantic-ai (PyPI) | 0.3.4 to before 1.107.4 | 1.107.4 | Patch available | GHSA-3gh4-cghq-f8v4 |
| — | pydantic-ai (PyPI) | 2.0.0b1 to before 2.27.1 | 2.27.1 | Patch available | GHSA-3gh4-cghq-f8v4 |
| — | pydantic-ai-slim (PyPI) | 0.3.4 to before 1.107.4 | 1.107.4 | Patch available | GHSA-3gh4-cghq-f8v4 |
| — | pydantic-ai-slim (PyPI) | 2.0.0b1 to before 2.27.1 | 2.27.1 | Patch available | GHSA-3gh4-cghq-f8v4 |
- pydantic-ai Vendorpydantic Affected 0.3.4 to before 1.107.4; 2.0.0b1 to before 2.27.1 Fixed in 1.107.4, 2.27.1 Patch available CNA
- pydantic-ai-slim Vendorpydantic Affected 0.3.4 to before 1.107.4; 2.0.0b1 to before 2.27.1 Fixed in 1.107.4, 2.27.1 Patch available CNA
- pydantic-ai (PyPI) Vendor— Affected 0.3.4 to before 1.107.4 Fixed in 1.107.4 Patch available GHSA-3gh4-cghq-f8v4
- pydantic-ai (PyPI) Vendor— Affected 2.0.0b1 to before 2.27.1 Fixed in 2.27.1 Patch available GHSA-3gh4-cghq-f8v4
- pydantic-ai-slim (PyPI) Vendor— Affected 0.3.4 to before 1.107.4 Fixed in 1.107.4 Patch available GHSA-3gh4-cghq-f8v4
- pydantic-ai-slim (PyPI) Vendor— Affected 2.0.0b1 to before 2.27.1 Fixed in 2.27.1 Patch available GHSA-3gh4-cghq-f8v4
Remediation
Update pydantic-ai and pydantic-ai-slim to 1.107.4 on the 1.x line or to 2.27.1 on the 2.x line. Red Hat lists Lightspeed Core and Red Hat Enterprise Linux AI (RHEL AI) 3 as affected with no fix plan.
Written with Claude Haiku 5.5 from the sources below · 2026-10-29
- Fix · CNA pydantic-ai (pydantic) Fixed in 1.107.4, 2.27.1
- Fix · CNA pydantic-ai-slim (pydantic) Fixed in 1.107.4, 2.27.1
- Fix · GHSA-3gh4-cghq-f8v4 pydantic-ai (PyPI) Fixed in 1.107.4 Advisory →
- Fix · GHSA-3gh4-cghq-f8v4 pydantic-ai (PyPI) Fixed in 2.27.1 Advisory →
- Fix · GHSA-3gh4-cghq-f8v4 pydantic-ai-slim (PyPI) Fixed in 1.107.4 Advisory →
- Fix · GHSA-3gh4-cghq-f8v4 pydantic-ai-slim (PyPI) Fixed in 2.27.1 Advisory →
Vendor assessments
Red Hat
Red Hat rating: Moderate. Red Hat's page for this CVE →
| Product | Status | Plan or fix |
|---|---|---|
| Lightspeed Core | Affected | No fix yet: Fix deferred |
| Red Hat Enterprise Linux AI (RHEL AI) 3 | Affected | No fix yet: Fix deferred |
Timeline
- 2026-10-08 Threatist record created from CVE-2026-107293
- 2026-10-08 Alias added: GHSA-3gh4-cghq-f8v4
- 2026-10-09 Affected versions changed · Red Hat
- 2026-10-09 CVSS score added · Red Hat
- 2026-10-13 CVSS score added · NVD
References
Patch 3
More Pydantic Ai vulnerabilities
- TI-2026-20687 Pydantic AI Affected by Server-Side Request Forgery (SSRF) in URL Download Handling CVSS 3.1HIGH 8.6
- TI-2026-69551 Pydantic AI keeps concurrency slots after streams end early CVSS 3.1HIGH 7.5
- TI-2026-46990 Incorrect authorization in pydantic-ai CVSS 3.1MEDIUM 6.5
- TI-2026-69610 Cross-site requests run agent tools in pydantic-ai web chat CVSS 3.1HIGH 7.6
- TI-2026-20740 Pydantic AI affected by Stored XSS via Path Traversal in Web UI CDN URL CVSS 3.1HIGH 7.1
- TI-2026-69617 Pydantic AI blocklist bypass via IPv6 zone identifier allows SSRF CVSS 3.1MEDIUM 6.8
Sources
- CVE-2026-107293 · CVE Program
- CVE-2026-107293 · NVD
- FIRST EPSS
- GHSA-3gh4-cghq-f8v4 · GitHub Advisory Database · CC BY 4.0
- CVE-2026-107293 · Red Hat security data · CC BY 4.0
This product uses the NVD API but is not endorsed or certified by the NVD. Sources and licences