CVE-2026-114386: Allocation of Resources Without Limits or Throttling in pydantic pydantic-ai-harness
Description
CVE-2026-114386 is a medium severity vulnerability in pydantic-ai-harness versions before 0.59.0. It involves the failure to enforce usage limits on nested tool calls within the CodeMode toolset, allowing users or malicious prompt content to bypass operator limits on resource-intensive or side-effecting tools.
CVSS v4.0
Score 6.3medium
Affected software
pydantic
pydantic-ai-harness
pkg:github/pydantic-ai-harnessRun on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
The vulnerability in pydantic-ai-harness prior to version 0.59.0 arises because the UsageLimits.tool_calls_limit is not enforced on nested tool calls dispatched from the run_code function in the CodeMode toolset. This allows users or prompt-injected content to invoke tools via run_nested_call() without triggering the check_before_tool_call validation, potentially leading to excessive resource consumption or unintended side effects by exceeding operator-defined limits on tool usage.
Potential Impact
Exploitation of this vulnerability can result in the sandboxed program invoking tools beyond the intended usage limits, which may cause resource exhaustion or unintended side effects. This could degrade system performance or lead to denial of service conditions. There is no indication of privilege escalation or data compromise from the provided data.
Mitigation Recommendations
A fix is available in pydantic-ai-harness version 0.59.0 and later. Users should upgrade to version 0.59.0 or above to ensure that UsageLimits.tool_calls_limit is properly enforced on nested tool calls. No additional mitigation steps are indicated in the vendor advisory.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulnCheck
- Date Reserved
- 2026-10-21T11:46:40.374Z
- Cvss Version
- 4.0
- State
- PUBLISHED
Threat ID: 6ad8b86a2cdf04f656f0ca53
Added to database: 10/21/2026, 13:04:42 UTC
Last enriched: 10/21/2026, 13:17:32 UTC
Last updated: 10/21/2026, 13:40:20 UTC
Views: 3
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.