Skip to content
Heist/ feed
Back to feed
Medium

MLflow CrewAI autologging records LLM provider API keys in logged trace spans

Type: advisorySources: Official CVE List

MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. From 2.19.0 until 3.17.0, mlflow.crewai.autolog() records CrewAI Agent objects without consistently sanitizing nested credential-bearing values. mlflow/crewai/autolog.py::_get_agent_attributes() converts the Agent llm field to a string that includes the provider API key, while mlflow/crewai/autolog.py::_construct_full_inputs() expands the Agent object into mlflow.spanInputs. Applications can therefore send plaintext provider credentials to the MLflow trace backend or a configured OTLP destination, where trace readers can obtain them. This issue is fixed in version 3.17.0.

Products
mlflow
Affected versions
Not specified
Fixed versions
Not specified
CVEs
CVE-2026-116691
Confirmed fixed CVEs
Not specified
Updated in this feed
31 Oct 2026, 15:42 UTC · Revision 1

Source evidence

MLflow CrewAI autologging records LLM provider API keys in logged trace spans

core-cve · Original source

Published 27 Oct 2026, 19:56 UTC

MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. From 2.19.0 until 3.17.0, mlflow.crewai.autolog() records CrewAI Agent objects without consistently sanitizing nested credential-bearing values. mlflow/crewai/autolog.py::_get_agent_attributes() converts the Agent llm field to a string that includes the provider API key, while mlflow/crewai/autolog.py::_construct_full_inputs() expands the Agent object into mlflow.spanInputs. Applications can therefore send plaintext provider credentials to the MLflow trace backend or a configured OTLP destination, where trace readers can obtain them. This issue is fixed in version 3.17.0.

cve:CVE-2026-116691

mlflow

Fixed versions: Not specified

Complete source record
{
  "id": "8009d3c52cfd60117dab619e323563d2a986816a472cce62a7803faf716dd358",
  "url": "https://www.cve.org/CVERecord?id=CVE-2026-116691",
  "method": "structured",
  "payload": {
    "kind": "advisory",
    "title": "MLflow CrewAI autologging records LLM provider API keys in logged trace spans",
    "summary": "MLflow is an open source AI engineering platform for agents, large language models, and machine learning models. From 2.19.0 until 3.17.0, mlflow.crewai.autolog() records CrewAI Agent objects without consistently sanitizing nested credential-bearing values. mlflow/crewai/autolog.py::_get_agent_attributes() converts the Agent llm field to a string that includes the provider API key, while mlflow/crewai/autolog.py::_construct_full_inputs() expands the Agent object into mlflow.spanInputs. Applications can therefore send plaintext provider credentials to the MLflow trace backend or a configured OTLP destination, where trace readers can obtain them. This issue is fixed in version 3.17.0.",
    "evidence": "cve:CVE-2026-116691",
    "products": [
      {
        "name": {
          "value": "mlflow",
          "evidence": "/containers/cna/affected/0/product"
        },
        "prerequisites": [],
        "fixed_versions": [],
        "affected_versions": []
      }
    ],
    "severity": "medium",
    "structured": {
      "id": "CVE-2026-116691",
      "aliases": [
        "CVE-2026-116691"
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "scope": "UNCHANGED",
            "version": "3.1",
            "baseScore": 6.5,
            "attackVector": "NETWORK",
            "baseSeverity": "MEDIUM",
            "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N",
            "integrityImpact": "NONE",
            "userInteraction": "NONE",
            "attackComplexity": "LOW",
            "availabilityImpact": "NONE",
            "privilegesRequired": "LOW",
            "confidentialityImpact": "HIGH"
          }
        }
      ],
      "affected": [
        {
          "vendor": "mlflow",
          "product": "mlflow",
          "versions": [
            {
              "status": "affected",
              "version": ">= 2.19.0, < 3.17.0"
            }
          ]
        }
      ],
      "modified": "2026-10-27T19:56:22.083Z",
      "provider": "cve",
      "references": [
        "https://github.com/mlflow/mlflow/security/advisories/GHSA-j7qq-vr8c-xvw3",
        "https://github.com/mlflow/mlflow/pull/25845",
        "https://github.com/mlflow/mlflow/commit/20927930d27f0072fcfeb7f024f24ad8c5e77c47",
        "https://github.com/mlflow/mlflow/releases/tag/v3.17.0"
      ],
      "provider_metadata": {
        "cna": {
          "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
          "shortName": "GitHub_M",
          "dateUpdated": "2026-10-27T19:56:22.083Z"
        }
      }
    },
    "related_cves": [
      {
        "value": "CVE-2026-116691",
        "evidence": "/cveMetadata/cveId"
      }
    ],
    "security_relevant": true,
    "severity_evidence": "/containers/cna/metrics"
  },
  "source_id": "core-cve",
  "document_id": 398117,
  "published_at": "2026-10-27T19:56:22.083+00:00"
}