Description
Pydantic AI Harness before 0.59.0 fails to enforce UsageLimits.tool_calls_limit on nested tool calls dispatched from run_code in the CodeMode toolset. Users or prompt-injected content can make the sandboxed program invoke tools via run_nested_call() without check_before_tool_call, exceeding operator limits on costly or side-effecting tools.
References6
- github.com/pydantic/pydantic-ai/blob/v2.58.0/src/pydantic_ai_ha…
- github.com/pydantic/pydantic-ai/commit/eb27b2b89fe76ae23ee9b6fc…
- github.com/pydantic/pydantic-ai/pull/10362
- github.com/pydantic/pydantic-ai/releases/tag/v2.59.0
- pypi.org/project/pydantic-ai-harness/0.59.0/
- www.vulncheck.com/advisories/pydantic-ai-harness-before-0.59.0-…
Metrics
- CVSS 3.1
-
4.8
MEDIUM
Score source: CNA
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:L/A:L - CVSS 4.0
-
6.3
MEDIUM
Score source: CNA
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:L/VA:L/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X - EPSS — exploit probability
- 0% percentile: 28 · as of 8 November 2026
- CWE
- CWE-770
Could this be exploited in your environment?
A patch closes the known hole. A penetration test shows what an attacker can still reach — with proof.