Authentication Bypass in Agno AGI Framework via Security Key Manipulation

A vulnerability was identified in agno-agi agno up to 2.7.3/3.0.6. This affects the function get_authentication_dependency of the file auth.py. Such manipulation of the argument security_key leads to missing authentication. The attack may be launched remotely. The exploit is publicly available and might be used. The project was informed of the problem early through an issue report but has not responded yet.

Casky was already ahead

This CVE exploits attack patterns that Casky's 283matched skills already investigate — long before this vulnerability was disclosed. Claude's reasoning model maps these techniques to MITRE ATT&CK, so practitioners who ran these skills have already seen the threat behaviour in their findings.

Analysis

CVE-2026-108818 represents a critical authentication bypass vulnerability in the Agno AGI framework (versions up to 2.7.3 and 3.0.6) where the get_authentication_dependency function in auth.py fails to properly validate the security_key parameter. This flaw allows attackers to circumvent authentication mechanisms entirely through remote exploitation, potentially granting unauthorized access to systems running vulnerable versions. Organizations deploying Agno for AI agent orchestration face immediate risk, particularly those exposing authentication endpoints to untrusted networks. The vulnerability is especially concerning given the public availability of exploit code and the framework maintainers' lack of response to early disclosure notifications.

Casky's 283 mapped skills detect the attack patterns underlying this CVE by correlating CLaude AI's extended reasoning with MITRE ATT&CK techniques TA0004 (Privilege Escalation) and TA0006 (Credential Access). Practitioners using Casky would observe detections flagging suspicious authentication flows where security_key parameters are being manipulated or omitted entirely, coupled with successful authentication outcomes that lack proper credential validation. The platform's skills would highlight anomalous patterns such as repeated failed authentication attempts suddenly succeeding, API calls bypassing expected security middleware, or session establishment without corresponding credential verification—all hallmarks of CWE-287 (Improper Authentication) and CWE-306 (Missing Authentication for Critical Function) exploitation.

Live Threat IntelligenceComing soon

Casky Risk Score

—

EPSS Probability

—

Shodan Exposure

—

GreyNoise Signal

—

Composite risk scoring from EPSS, CISA KEV, Shodan, and GreyNoise — 21 security APIs correlated into a single Casky Risk Score. Coming in Casky Pro. Join early access →

283 Casky skills cover this attack pattern

These skills use Claude AI's reasoning model to surface findings in the same attack categories as CVE-2026-108818.

abusing-dpapi-for-credential-access

red teaming · high

Run

abusing-shadow-credentials-for-privesc

red teaming · high

Run

Access with Stolen Session Cookie

identity access management · low

Run

MITRE ATT&CK Techniques

TA0004 TA0006

Investigate the attack patterns behind CVE-2026-108818

Casky has 283 skills that investigate the attack patterns behind CVE-2026-108818. Run one and get CVSS-scored findings in 3 minutes.

Run the skill that detects this →

X Instagram LinkedIn