PT-2026-60412

The BetterDocs WordPress plugin before 4.5.5 does not sanitise an AI-generated documentation summary before storing and outputting it, and the feature that generates it is exposed to unauthenticated users, allowing them to store a malicious payload via prompt injection that executes in the browse...

CVE-2026-11371: Improper Neutralization of Input During Web Page Generation

The BetterDocs WordPress plugin before 4.5.5 does not sanitise an AI-generated documentation summary before storing and outputting it, and the feature that generates it is exposed to unauthenticated users, allowing them to store a malicious payload via prompt injection that executes in the browse...

PT-2026-60484

Name of the Vulnerable Software and Affected Versions @prompty/core versions 2.0.0-alpha.1 through 2.0.0-beta.2 Description The @prompty/core TypeScript loader in runtime/typescript/packages/core/src/core/loader.ts uses gray-matter without overriding executable js and javascript frontmatter...

CVE-2026-58658

Summary. CVE-2026-58658 affects GPUStack up to version 2.2.1 and was fixed by commit 4e20551. An unauthenticated attacker can abuse unprotected /serveLogs and /debug endpoints on the worker port to disclose sensitive information and configure logs. Specifically, attackers can enumerate model inst...

How Agents Ask for Permission: User Permissions for AI Agents, from Interfaces to Enforcement

As AI agents gain prevalance, users are increasingly exposed to the risks such systems entail. Prompt injection attacks, as well as hallucination, can cause agents to leak private information to third parties. As autonomous systems, agents also present the more active danger of performing sensiti...

Context Contamination in LLM Analysis of Network Security Logs: Poison with Passive Prompt Injection and Mitigation Evaluation

Large Language Models are increasingly deployed in Security Operations Centers for log analysis tasks including summarization, alert triage, and threat investigation. These systems ingest logs from external-facing services and process network logs as natural language contexts to generate security...

Rethinking Penetration Testing for AI-Enabled Systems: From Resource Compromise to Behavioral Objective Violation

Penetration testing traditionally evaluates whether adversaries can exploit weaknesses in software, infrastructure, configurations, or operational controls to achieve security-relevant compromise. This paradigm remains necessary for AI-enabled systems, but it is no longer sufficient. In such...

Researchers Say Claude for Chrome Flaw Lets Rogue Extensions Trigger Gmail Reads

Any other browser extension that can run a script on claude.ai can still trigger Claude for Chrome tasks aimed at your Gmail, your latest Google Doc and its comments, and your Calendar. Both this and ClaudeBleed need a rogue extension that can already run a script on claude.ai; the difference is...

Langroid: Sandbox Escape to Remote Code Execution via Incomplete `eval()` Mitigation in TableChatAgent

Advisory DetailsTitle: Sandbox Escape to Remote Code Execution via Incomplete eval Mitigation in TableChatAgentDescription: SummaryLangroid is vulnerable to a critical Sandbox Escape leading to Remote Code Execution RCE in its TableChatAgent and VectorStore capabilities. When these agents evaluat...

Langroid: Neo4jChatAgent executes LLM-generated Cypher without validation (prompt-to-Cypher injection; config-conditional RCE), mirroring the SQLChatAgent bug fixed in CVE-2026-25879

Neo4jChatAgent passes LLM-generated Cypher queries straight to the Neo4j driver with no validation, no statement-type allowlist, and no opt-out gate. The query text is influenceable by prompt injection direct user input or indirect content the agent reads back via RAG, so an attacker who can...

PYSEC-2026-2576 Langroid: Neo4jChatAgent executes LLM-generated Cypher without validation (prompt-to-Cypher injection; config-conditional RCE), mirroring the SQLChatAgent bug fixed in CVE-2026-25879

Neo4jChatAgent passes LLM-generated Cypher queries straight to the Neo4j driver with no validation, no statement-type allowlist, and no opt-out gate. The query text is influenceable by prompt injection direct user input or indirect content the agent reads back via RAG, so an attacker who can...

Open WebUI: Cross-origin postMessage confirmation bypass via action:submit

SummaryThe chat message listener allows non-same-origin input:prompt and action:submit messages, so an external site can set prompt text and trigger submitPrompt in an authenticated victim session. I validated this with a cross-origin attacker page that auto-posted messages and caused unauthorize...

Open WebUI Prompt history IDOR: unbound history_id allows cross-prompt read and deletion

SummaryOpen WebUI's prompt version-history endpoints authorize the promptid in the URL but then act on caller-supplied history IDs without verifying that the history row belongs to that prompt historyentry.promptid == prompt.id. Three operations are affected:- GET...

PYSEC-2026-2701 Open WebUI Prompt history IDOR: unbound history_id allows cross-prompt read and deletion

Summary Open WebUI's prompt version-history endpoints authorize the promptid in the URL but then act on caller-supplied history IDs without verifying that the history row belongs to that prompt historyentry.promptid == prompt.id. Three operations are affected: - GET...

PYSEC-2026-2903 PraisonAI CLI automatically resolves @url mentions in prompt text and can read loopback URLs into model context

Summary PraisonAI's direct-prompt CLI automatically expands @url: mentions in raw prompt text before agent execution begins. If a prompt contains @url:, the CLI calls MentionsParser.process.... The @url: handler then performs a direct urllib.request.urlopen request to the attacker-controlled URL...

PraisonAI CLI automatically resolves @url mentions in prompt text and can read loopback URLs into model context

SummaryPraisonAI's direct-prompt CLI automatically expands @url: mentions in raw prompt text before agent execution begins.If a prompt contains @url:, the CLI calls MentionsParser.process.... The @url: handler then performs a direct urllib.request.urlopen request to the attacker-controlled URL an...

PraisonAI CLI automatically resolves @url mentions in prompt text and can read loopback URLs into model context

SummaryPraisonAI's direct-prompt CLI automatically expands @url: mentions in raw prompt text before agent execution begins.If a prompt contains @url:, the CLI calls MentionsParser.process.... The @url: handler then performs a direct urllib.request.urlopen request to the attacker-controlled URL an...

PYSEC-2026-2942 PraisonAI CLI automatically resolves @url mentions in prompt text and can read loopback URLs into model context

Summary PraisonAI's direct-prompt CLI automatically expands @url: mentions in raw prompt text before agent execution begins. If a prompt contains @url:, the CLI calls MentionsParser.process.... The @url: handler then performs a direct urllib.request.urlopen request to the attacker-controlled URL...

PYSEC-2026-2514 hermes-agent has an Injection issue

A vulnerability was found in NousResearch hermes-agent 2026.4.23. The impacted element is the function scancontextcontent of the file agent/promptbuilder.py. The manipulation results in injection. The attack may be performed from remote. The exploit has been made public and could be used. The...

PYSEC-2026-2509 hermes-agent has an Injection issue

A security vulnerability has been detected in NousResearch hermes-agent up to 2026.4.23. Impacted is an unknown function of the file agent/skillsguard.py of the component Skills Guard Multi-Word Prompt Handler. The manipulation of the argument THREATPATTERNS leads to injection. Remote exploitatio...