CVE-2026-57118

validatedZero produced an exploit or a check.

GHSA-x8cv-xmq7-p8xp ↗EUVD-2026-97774 ↗

check availableZero validated a working check for it.not proven exploitableZero has a detection check, but hasn't produced a full working exploit yet.
cve.orgpublic pocExploit code has been published by someone. Source signals can also carry a PoC (Proof of Concept).−117dkevAttacks confirmed in the real world.
MervinPraison PraisonAI <= 4.6.57
fixed in >= 1.6.59
CWE-306 Missing Authentication for Critical Function
ETK-2026-1406Zero's internal tracking ID, independent of the CVE number., permanent identifier
pkg:pypi/praisonaiagents
+44mTTV
15mAI analysisAnalysis: how long the AI spent reversing this signal.
Check if you are vulnerable
timeline
0 = cve.org publication
public poc−3.9motrigger signal (cve.org) +0mocheck available+0.0moagent stopped
−3mo
−2mo
−1mo

description

PraisonAI AgentTeam.launch() and its Agents alias can expose HTTP endpoints that do not require authentication before listing or invoking agents. An attacker who can reach a launched server may enumerate agents and cause agent operations, including consumption of model resources or actions through configured tools and context.

mitigation

Upgrade PraisonAI to 4.6.59 and praisonaiagents to 1.6.59 or later. Until upgrades are complete, do not expose AgentTeam.launch() servers to untrusted networks; bind them to a local or otherwise access-restricted interface.

fixed in >= 1.6.59

check
available
exploit
no
automationDeterministic: a check that runs on its own, the same request on every target. Contextual: an agentic technique that needs to be applied per target, for bugs that depend on contextual flows.
contextual
auth requiredWhether an attacker needs an account first, based on the CVSS vector's Privileges Required.
no
EPSSEstimate of the chance this CVE is exploited in the next 30 days. Refreshed daily.
0.33%24th percentile
agent run time
15m
cve.org record, t = 0The moment cve.org publishes the CVE in full.
2026-10-12 18:48Z
Δ to public pocExploit code has been published by someone. Source signals can also carry a PoC (Proof of Concept).
−117d
Δ to kevAttacks confirmed in the real world.
–

outcome

Mechanism verified: the exploit fired on the vulnerable build and stayed silent on the patched one, but the server it ran against was ours, so no check ships for it.

source signals

Public signals that triggered or updated the pipeline. These are inputs for the deltas above.

  1. 2026-06-17 07:32Z
  2. 2026-10-12 18:48Z
    cve_org: trigger signal, started the run

this entry in the feed

GET /api/v1/feed/CVE-2026-57118

{
  "etk_id": "ETK-2026-1406",
  "cve_id": "CVE-2026-57118",
  "display_id": "CVE-2026-57118",
  "ghsa_id": "GHSA-x8cv-xmq7-p8xp",
  "euvd_id": "EUVD-2026-97774",
  "status": "published",
  "anchor_state": "anchored",
  "title": "Unauthenticated agent invocation in PraisonAI AgentTeam.launch()",
  "summary": "PraisonAI AgentTeam.launch() and its Agents alias can expose HTTP endpoints that do not require authentication before listing or invoking agents. An attacker who can reach a launched server may enumerate agents and cause agent operations, including consumption of model resources or actions through configured tools and context.",
  "mitigation": "Upgrade PraisonAI to 4.6.59 and praisonaiagents to 1.6.59 or later. Until upgrades are complete, do not expose AgentTeam.launch() servers to untrusted networks; bind them to a local or otherwise access-restricted interface.",
  "vendor": "MervinPraison",
  "product": "PraisonAI",
  "affected": [
    {
      "affected_notes": [],
      "affected_ranges": [
        "<= 4.6.57"
      ],
      "ecosystem": "pip",
      "fixed_versions": [
        ">= 1.6.59"
      ],
      "package_name": "praisonaiagents",
      "purl": "pkg:pypi/praisonaiagents"
    }
  ],
  "cwes": [
    306
  ],
  "cvss_score": 9.8,
  "cvss_src": "cna_vendor",
  "cvss_vector": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
  "epss_score": 0.0033,
  "epss_percentile": 0.2415,
  "epss_asof": "2026-10-20",
  "automation": "contextual",
  "check": true,
  "check_confidence_basis": "mechanism_verified",
  "exploit": false,
  "pipeline_minutes": 15.3204644,
  "t0_at": "2026-10-12T18:48:22.664000+00:00",
  "signals": {
    "check": {
      "at": "2026-10-12T19:32:20.994673+00:00",
      "days": 0.030536234641203704,
      "is_lower_bound": false,
      "measured_at": "2026-10-12T18:48:22.664000+00:00",
      "url": null
    },
    "public_poc": {
      "at": "2026-06-17T07:32:44+00:00",
      "days": -117.4691975,
      "is_lower_bound": false,
      "measured_at": "2026-10-12T18:48:22.664000+00:00",
      "url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-x8cv-xmq7-p8xp"
    },
    "trigger_signal": {
      "at": "2026-10-12T18:48:22.664000+00:00",
      "days": 0,
      "is_lower_bound": false,
      "measured_at": "2026-10-12T18:48:22.664000+00:00",
      "url": "https://www.cve.org/CVERecord?id=CVE-2026-57118"
    }
  }
}

published 2026-10-12 19:16Z
last change 2026-10-21 09:43Z

Need the technical details behind an entry?

Every entry here shows what we attempted and when. The technical details, the generated exploit that fired it and the evidence behind the verdict are not public. Tell us about your use case and we’ll get back to you about extended API access.