Xinference: Remote code execution via unsafe `eval()` in Llama3 tool-call parsing
CVSS score
10.0
EPSS probability
%1.2
Risk score
0.1
Published
1 mo ago
Etkilenen ürün ve sürüm için yayımlanan güvenlik güncellemesini uygulayın.
Impact on My Inventory
Only the personal products you have added to your account are assessed; an organisation's inventory does not appear here.
Checking session…
Vulnerability description
DayBreach CVE AI Research
CVE-2026-61539 has vendor, distribution, patch and security-source records on file. We can compare them and prepare an explanation linked to the evidence.
Xinference is an inference API for running open-source, speech, and multimodal models. In 2.5.0 and earlier, Xinference passes attacker-influenced Llama3 tool-call output to eval() in xinference/model/llm/tool_parsers/llama3_tool_parser.py and xinference/model/llm/utils.py. Requests to /v1/chat/completions with a tools field flow through xinference/api/restful_api.py, xinference/model/llm/transformers/core.py, handle_chat_result_non_streaming(), and _post_process_completion() before extract_tool_calls() or _eval_llama3_chat_arguments() evaluates the model-generated Python expression. An unauthenticated remote attacker can influence that output through a crafted prompt and execute commands in the Xinference server process context. This issue is fixed in version 2.7.0.
IMPACT: Affected Products and Software
RED HAT CSAF/VEX: Product Impact Check
SCORING: CVSS from Multiple Sources
CVSS Scores
The Common Vulnerability Scoring System is a standardised framework for assessing the severity of vulnerabilities in software and systems. We collect and display CVSS scores from several sources for each CVE.
| SCORE | VERSION | SEVERITY | VECTOR | SOURCE |
|---|---|---|---|---|
| 10.0 | CVSS 3.x | — | Vector not published | GitHub Advisory |
| 10.0 | CVSS V3_1 | CRITICAL | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H | CVE CNA |
| 10.0 | CVSS V31 | CRITICAL | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H | NVD |
| 0.0 | CVSS 4.0 | — | Vector not published | GitHub Advisory |
Patch and Remediation Guide
Solution & Remediation Advisory
1 doğrulanmış yama veya düzeltilmiş sürüm kanıtı bulundu.
2.7.0OSV düzeltilmiş sürüm: 2.7.0
Public exploit and active exploitation status
Verified exploit intelligence
No verified public PoC, exploit repository or Metasploit module has been found for this CVE yet.
Weakness Classification and Attack Patterns (CWE & CAPEC)
CWE - Common Weakness Enumeration
While CVE identifies specific instances of vulnerabilities, CWE categorises the common flaws or weaknesses that can lead to vulnerabilities. CVE-2026-61539 is associated with the following CWEs:
Common Attack Pattern Enumeration and Classification (CAPEC)
Common Attack Pattern Enumeration and Classification (CAPEC) stores attack patterns, which are descriptions of the common attributes and approaches employed by adversaries to exploit the CVE-2026-61539 weaknesses.
Attack Vector and Accessibility
Official Sources and Advisories
CVSS Vector Radar Chart
10.0 / 10Vulnerability Scoring Details
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:HYama doğrulandı
Weakness Type (CWE & CAPEC)
Similar Vulnerabilities
MLflow: Unauthenticated full-read SSRF in webhook delivery: _validate_webhook_url bypassed via unvalidated HTTP redirects (and DNS rebinding)
SQL injection in Litellm
netlicensing-mcp: REST Path Traversal Bypasses Token Redaction
Starlette has missing Host header validation that poisons request.url.path, bypassing path-based security checks
ESPHome Device Builder Dashboard: Unauthenticated dashboard access via the HA add-on ingress site bound to all interfaces
Natural Language Toolkit (NLTK): DNS-rebinding SSRF filter bypass in nltk.pathsec.urlopen (nltk.download / nltk.data.load) defeats ENFORCE mode
Get this vulnerability through the API
Affected products, version ranges, exploitation status and patch information in a single request, with a free key.
curl -H "x-api-key: $DAYBREACH_API_KEY" \ "https://api.enginteksut.com.tr/api/v1/intel/cves/CVE-2026-61539"