☠️
EXFILTRATING
DATA...
01010111 10110010 00101001 11000110 01101100 10011011 00010001 11100101 01000010 10101111
[ <-- BACK_TO_SEARCH ]
Tencent logo [ WEBSITE ] [ VENDOR STATS ]
CVSS: 6.9 🐛 CWE-200
👁️
Common Weakness Enumeration - Identifies the specific category of software weakness or flaw.

CVE-2026-5585

CVE-2026-5585 is an information disclosure vulnerability in Tencent AI-Infra-Guard version 4.
📅 PUBLISHED: 2026-04-05 17:30 UTC 🔄 UPDATED: 2026-08-18 17:14 UTC
[ ZERO-DAY / UNPATCHED ] [ DEFENSE PATCH ↓ ]
💣 PoC: None
👁️
Proof of Concept - Indicates if functional exploit code is publicly available, drastically increasing the risk of immediate attacks.
🎯 EPSS: 0.64%
👁️
0.64% probability
16.0× above the median CVE
Tier: ELEVATED
30-day exploitation window · FIRST.org · updated daily
🥷 MITRE: T1552
👁️
MITRE ATT&CK Tactic - Describes the strategy hackers use to execute this attack.
VULNERABLE: UNKNOWN / CHECK VENDOR
CVSS SEVERITY SCALE
9.0–10.0 CRITICAL
7.0–8.9 HIGH
4.0–6.9 MEDIUM
0.1–3.9 LOW
0.0 NONE
6.9
THREAT LEVEL
■ tracking this CVE ✓ independent patch confirmation ⧗ same-origin mirror (GHSA) ⚠ active exploitation verification policy →

Executive Summary

CVE-2026-5585 is an information disclosure vulnerability in Tencent AI-Infra-Guard version 4.0, specifically within the Task Detail Endpoint implemented in common/websocket/task_manager.go. The flaw allows an unauthenticated remote attacker to access sensitive task details without proper authorization. The exploit has been publicly released, and the vendor has not responded to disclosure attempts. With a CVSS score of 5.3 (Medium), the risk is elevated due to the lack of authentication requirements and the public availability of exploit code.

⚡ AUTOMATE YOUR CVE INTELLIGENCE

This record — and 170,000+ others — are available via the ValtersIT API. One endpoint replaces five: NVD + GitHub CVE + CISA KEV + EPSS + Nuclei + Exploit-DB, unified.

›Patch confidence — UNVERIFIED to CONFIRMED, cross-referenced NVD + RHSA + Debian
›Exploit & PoC — poc_status, poc_url, has_exploit. Not in NVD.
›Shodan dorks — ready-made queries to find exposed vulnerable systems
›Detection rules — YARA · Wazuh · Elastic (all plans) · Sigma (Pro)
›pending_fields — know exactly which of 7 fields NVD hasn't processed yet
›Free backfill — data incomplete when you paid? Re-check within 30 days = 0 credits
PAYG €0.08/lookup Standard €19/mo (was €29) · 5k credits Pro €65/mo (was €99) · 20k credits FOUNDING PRICE · LOCKED FOR LIFE · ENDS 31 DEC 2026
[ FROM €19/MO ]
Free sandbox key available
No signup required

Supply Chain & Package Security

This vulnerability affects a Go-based application (Tencent AI-Infra-Guard). The project is not distributed via standard package registries (NPM, PyPI, RubyGems), but organizations using this tool should verify their deployment source. Since no patch exists, users must rely on manual mitigation or consider removing the component from production environments.

CVSS Breakdown & Threat Level

  • Attack Vector (AV): Network (N) – Exploitable remotely.
  • Attack Complexity (AC): Low (L) – No special conditions required.
  • Privileges Required (PR): None (N) – No authentication needed.
  • User Interaction (UI): None (N) – No user action required.
  • Scope (S): Unchanged (U) – Impact confined to the vulnerable component.
  • Confidentiality (C): Low (L) – Limited information disclosure.
  • Integrity (I): None (N) – No data modification.
  • Availability (A): None (N) – No denial of service.

Threat Level: Medium. The low EPSS score (0.0%) suggests limited active exploitation, but the public exploit and lack of patch increase the urgency for detection and mitigation.

Deep Dive & Mechanics

The vulnerability resides in the WebSocket-based Task Detail Endpoint within common/websocket/task_manager.go. The endpoint fails to enforce proper access controls, allowing any remote client to query task details without authentication. The manipulation likely involves sending a crafted WebSocket message to retrieve task metadata, including potentially sensitive configuration or operational data.

The root cause is a missing authorization check in the WebSocket handler. The function processes incoming requests and returns task details without verifying the requester’s identity or permissions. This is a classic case of CWE-200 (Exposure of Sensitive Information to an Unauthorized Actor).

Vulnerable vs. Patched Code

Since no patch is available, the vulnerable code is the only reference. The following snippet illustrates the likely vulnerable pattern in common/websocket/task_manager.go:

// Vulnerable code (CVE-2026-5585)
func handleTaskDetail(conn *websocket.Conn, msg map[string]interface{}) {
    taskID := msg["task_id"].(string)
    // Missing authentication check
    taskData := getTaskDetails(taskID) // Returns sensitive data
    conn.WriteJSON(taskData)
}

A patched version would include an authentication check:

// Hypothetical patched code
func handleTaskDetail(conn *websocket.Conn, msg map[string]interface{}) {
    if !isAuthenticated(conn) {
        conn.WriteJSON(map[string]string{"error": "unauthorized"})
        return
    }
    taskID := msg["task_id"].(string)
    taskData := getTaskDetails(taskID)
    conn.WriteJSON(taskData)
}

Exploitation & Lab Setup

Red Team

Prerequisites:

  • Target running Tencent AI-Infra-Guard 4.0 with WebSocket endpoint exposed.
  • Network access to the target (e.g., ws://target:8080/ws).

Exploit Steps:

  1. Connect to the WebSocket endpoint.
  2. Send a JSON message requesting task details:
    {"action": "task_detail", "task_id": "any_task_id"}
    
  3. Receive the response containing sensitive task information.

Example using websocat:

websocat ws://target:8080/ws
> {"action":"task_detail","task_id":"1"}
< {"task_id":"1","status":"running","config":"sensitive_data"}

Blue Team

Detection:

  • Monitor WebSocket connections to the /ws endpoint.
  • Alert on requests containing "action":"task_detail" without prior authentication.
  • Use the Sigma rule below to detect anomalous WebSocket messages.

Mitigation:

  • Restrict network access to the WebSocket endpoint using firewall rules.
  • Implement authentication middleware for all WebSocket handlers.
  • If possible, disable the Task Detail Endpoint until a patch is available.

Indicators of Compromise (IOCs)

  • Network IOCs:

    • WebSocket connections to ws://<target>:8080/ws with JSON payloads containing "action":"task_detail".
    • Outbound connections from the target to unknown IPs (if data exfiltration occurs).
  • File IOCs:

    • Log entries in task_manager.log showing unauthenticated task detail requests.
    • Modified configuration files in /etc/ai-infra-guard/ (if exploited for further access).
  • Behavioral IOCs:

    • Unusual number of WebSocket connections from a single IP.
    • Requests for task IDs that do not correspond to legitimate tasks.

Detection Engineering

YARA Rule

rule CVE_2026_5585_WebSocket_Exploit {
    meta:
        description = "Detects exploit attempts for CVE-2026-5585 in Tencent AI-Infra-Guard"
        author = "Valters IT"
        date = "2025-04-10"
        reference = "https://nvd.nist.gov/vuln/detail/CVE-2026-5585"
    strings:
        $ws_payload = /"action"\s*:\s*"task_detail"/
        $task_id = /"task_id"\s*:\s*"[^"]+"/
    condition:
        $ws_payload and $task_id
}

Sigma Rule

title: CVE-2026-5585 WebSocket Task Detail Exploit
id: 8a4b3c2d-1e5f-4a7b-9c8d-0e1f2a3b4c5d
status: experimental
description: Detects unauthenticated WebSocket requests to the Task Detail endpoint
author: Valters IT
logsource:
    category: websocket
    product: tencent_ai_infra_guard
detection:
    selection:
        uri: "/ws"
        payload|contains: "task_detail"
    condition: selection
falsepositives:
    - Legitimate internal monitoring tools (if authenticated)
level: high

Suricata Rule

alert tcp any any -> any 8080 (msg:"CVE-2026-5585 WebSocket Task Detail Exploit"; flow:to_server; content:"|7b 22 61 63 74 69 6f 6e 22 3a 22 74 61 73 6b 5f 64 65 74 61 69 6c 22|"; classtype:attempted-recon; sid:1000001; rev:1;)

Remediation & Mitigation

Immediate Actions:

  1. Isolate the vulnerable component from untrusted networks (e.g., place behind a VPN or firewall).
  2. Monitor WebSocket traffic for exploitation attempts using the provided detection rules.
  3. Disable the Task Detail Endpoint if not required, by removing or commenting out the WebSocket handler in common/websocket/task_manager.go.

Long-term Fix:

  • Since the vendor is unresponsive, consider forking the repository and implementing authentication checks as shown in the patched code example.
  • Evaluate alternative tools that provide similar functionality with active security support.

Workaround:

  • Implement a reverse proxy (e.g., Nginx) that requires HTTP Basic Authentication before proxying WebSocket connections to the backend.

Valters IT

{/* [Security Hash: C9BA224B74E6F417165621F1] */}

◈ SOURCE Analysis & enrichment: valtersit.com/cve/CVE-2026-5585/ · CVE data: NVD/MITRE (public domain) · © 2026 ValtersIT Hub
>_ SYSTEM_CORE: INTEL_ACTIVE
90% TEAM-CURATED // 10% AUTOMATED INGESTION.
ALL PATCH REFERENCES AND VERSION DATA SOURCED FROM NVD + OSV OFFICIAL REGISTRIES.