PT-2026-78822

Name of the Vulnerable Software and Affected Versions Splunk AI Toolkit versions prior to 6.0.0 Description A user with the "power" Splunk role can execute arbitrary code on the Splunk server by loading a model file containing crafted sparse matrix data. This occurs because a model codec...

PT-2026-78818

Name of the Vulnerable Software and Affected Versions Splunk AI Toolkit versions prior to 6.0.0 Description Improper privilege management in the Agent Run History handler allows users without admin or power roles to execute searches with system-level privileges. This occurs because the handler...

PT-2026-78823

Name of the Vulnerable Software and Affected Versions Splunk AI Toolkit versions prior to 6.0.0 Description Improper access control allows a user with the schedule search capability to cause a scheduled search to load and deserialize a model file. This occurs because the apply search command is n...

PT-2026-78820

Name of the Vulnerable Software and Affected Versions Splunk AI Toolkit versions prior to 6.0.0 Description A race condition exists when users upload models. An attacker with upload permissions can overwrite a model being uploaded by another user by sending a concurrent upload request using the...

PT-2026-78819

🚨 CVE-2026-76392 In Splunk AI Toolkit versions below 6.0.0, a user who does not hold the "admin" or "power" Splunk roles could obtain predictable or default credentials for connected container services. The use of hard-coded credentials is possible because Splunk AI Toolkit generates or stores...

PT-2026-78821

Name of the Vulnerable Software and Affected Versions Splunk AI Toolkit versions prior to 6.0.0 Description Low-privileged users without admin or power roles can start, stop, and configure containers, as well as read or modify connection and configuration data. This occurs because multiple handle...

EUVD-2026-61423

Vulnerability in the Siebel Artificial Intelligence product of Oracle Siebel CRM component: AI. Supported versions that are affected are 25.12-26.6. Easily exploitable vulnerability allows low privileged attacker with network access via HTTP to compromise Siebel Artificial Intelligence. While the...

CVE-2026-75130

Context7 through 2.1.2 contains a prompt injection vulnerability that allows attackers to execute malicious instructions in connected AI coding agents by injecting unsanitized content through the Custom AI Instructions feature served via the MCP server. Attackers can poison the custom instruction...

CVE-2026-75130

Context7 through 2.1.2 has a prompt injection flaw in its Custom AI Instructions feature served via the MCP server. Attackers poison unsanitized instructions to exfiltrate credentials from environment files and delete files on victim machines when the agent makes routine documentation requests. C...

CVE-2026-75130 Context7 2.1.2 Prompt Injection via Custom AI Instructions

Context7 through 2.1.2 contains a prompt injection vulnerability that allows attackers to execute malicious instructions in connected AI coding agents by injecting unsanitized content through the Custom AI Instructions feature served via the MCP server. Attackers can poison the custom instruction...

CVE-2026-75859

CodeWhale versions before 0.8.64 fail to validate file paths in the project config instructions field, allowing attackers to read arbitrary files on the victim's system. A malicious .codewhale/config.toml file in a cloned repository can specify paths outside the workspace that are read and inject...

CVE-2026-75911

CodeWhale versions before 0.8.64 fail to properly validate the allowshell configuration parameter from project config files, allowing attackers to enable arbitrary shell command execution by committing a malicious .codewhale/config.toml file to a repository. When a user clones and opens the...

EUVD-2026-60966

CodeWhale versions before 0.8.64 fail to validate file paths in the project config instructions field, allowing attackers to read arbitrary files on the victim's system. A malicious .codewhale/config.toml file in a cloned repository can specify paths outside the workspace that are read and inject...

CVE-2026-75859

CodeWhale before 0.8.64 fails to validate file paths in the project config instructions field. A malicious .codewhale/config.toml in a cloned repo can read files outside the workspace and inject their contents into the AI system prompt, enabling exfiltration. CVSS 7.5 HIGH. Patch released; no kno...

CVE-2026-75859 CodeWhale before 0.8.64 Arbitrary File Read via instructions

CodeWhale versions before 0.8.64 fail to validate file paths in the project config instructions field, allowing attackers to read arbitrary files on the victim's system. A malicious .codewhale/config.toml file in a cloned repository can specify paths outside the workspace that are read and inject...

CVE-2026-75859

CodeWhale versions before 0.8.64 fail to validate file paths in the project config instructions field, allowing attackers to read arbitrary files on the victim's system. A malicious .codewhale/config.toml file in a cloned repository can specify paths outside the workspace that are read and inject...

CVE-2026-75859 CodeWhale before 0.8.64 Arbitrary File Read via instructions

CodeWhale versions before 0.8.64 fail to validate file paths in the project config instructions field, allowing attackers to read arbitrary files on the victim's system. A malicious .codewhale/config.toml file in a cloned repository can specify paths outside the workspace that are read and inject...

ANT-2026-AM0R21ZY · supabase/supabase · broken-access-control

broken-access-control medium Severity Claude medium · Security research firm - · Maintainer - Discovered by Claude Mythos Preview REPORT Anthropic's analysis of this finding, sealed at approval. ANT-2026-AM0R21ZY: HIPAA sensitive-project AI block enforced only client-side The client hook...

ANT-2026-Y9HESEM8 · supabase/supabase · auth-bypass

auth-bypass medium Severity Claude medium · Security research firm - · Maintainer - Discovered by Claude Mythos Preview REPORT Anthropic's analysis of this finding, sealed at approval. ANT-2026-Y9HESEM8: AI data-sharing opt-in bypassed by mismatched org and project In...

AI "Mind Viruses" Can Spread Between Agents Through Persistent Prompt Files

Security researchers at Anthropic and Switzerland's EPFL have demonstrated that self-propagating payloads can spread from one artificial intelligence AI agent to the next through the editable system prompt files that autonomous agent harnesses use to carry state between sessions. The work, releas...