PT-2026-61797

Name of the Vulnerable Software and Affected Versions praisonaiagents affected versions not specified Description A Server-Side Request Forgery SSRF issue exists in the SearxNG and search web search tools. The searxng url variable is passed to the requests.get function without validation of the...

PT-2026-50689

Name of the Vulnerable Software and Affected Versions Eclipse Theia versions prior to 1.71.0 Description The AI chat agent processes workspace file and directory names as part of its prompt context without distinguishing them from system instructions. This allows for indirect prompt injection,...

PT-2026-61739

Name of the Vulnerable Software and Affected Versions @agenticmail/claudecode versions prior to 0.2.39 @agenticmail/codex versions prior to 0.1.33 @agenticmail/core versions prior to 0.9.43 @agenticmail/openclaw versions prior to 0.5.71 Description Two inbound-mail handlers execute privileged...

CVE-2026-22551: Insertion of Sensitive Information Into Sent Data

In Eclipse Theia versions prior to 1.71.0, the AI chat rendered Markdown image tags from AI responses, triggering HTTP requests to arbitrary external URLs without restriction. Combined with prompt injection in a malicious workspace, an attacker could induce the AI agent to construct image URLs...

GHSA-4R4W-2WGP-W7CJ 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...

Authorization Bypass Through User-Controlled Key

Overview open-webui is an Open WebUI Affected versions of this package are vulnerable to Authorization Bypass Through User-Controlled Key in the handling of prompt history operations, including computediff, updatepromptversion, and deletehistoryentry. An attacker can access or delete another user...

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...

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

Summary The 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 unauthoriz...

GHSA-3VV5-8XXP-4F55 Open WebUI: Cross-origin postMessage confirmation bypass via action:submit

Summary The 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 unauthoriz...

CVE-2026-55743 OpenHuman desktop agent shell tool sandbox bypass leads to arbitrary command execution

The shell tool command allowlist in the SecurityPolicy of OpenHuman desktop agent through 0.54.0 default Supervised security policy can be bypassed to execute arbitrary OS commands with the privileges of the desktop user...

CVE-2026-55743

OpenHuman desktop agent (before 0.54.0, fixed in 0.56.0) contains two policy flaws in src/openhuman/security/policy.rs that bypass the shell allowlist, enabling remote code execution via indirect prompt injection. First, is_args_safe() blocks -exec and -ok while not blocking -execdir/-okdir (whic...

GHSA-MQXH-6GQ7-558M Pi Agent: Pi loads project-local extensions without approval

Pi loads project-local extensions without approval Pi before 0.79.0 loaded project-local configuration and resources from a repository's .pi directory without first asking the user to trust that repository. This included project-local extensions, which are executable TypeScript or JavaScript...

PT-2026-50479

Name of the Vulnerable Software and Affected Versions Open WebUI versions prior to 0.9.6 Description The chat message listener in the chat page's window message listener processes input:prompt and action:submit messages without enforcing same-origin restrictions. This allows an external site to s...

A Layered Security Framework against Prompt Injection in RAG-Based Chatbots

Prompt injection is ranked as the most critical vulnerability in large language model LLM deployments by the OWASP Top 10 for LLM Applications, yet existing defenses operate at isolated pipeline stages and remain incomplete. Input filters cannot inspect retrieved documents, while output monitors...

PT-2026-50487

Name of the Vulnerable Software and Affected Versions Open WebUI versions prior to 0.9.6 Description Open WebUI contains an authorization flaw in its prompt version-history endpoints. While the system authorizes the prompt id provided in the URL, it fails to verify that the requested history...

PT-2026-50492

Name of the Vulnerable Software and Affected Versions Pi versions prior to 0.79.0 Description Pi loads project-local configuration and resources from a repository's .pi directory without requiring user trust. This includes project-local extensions, which are executable TypeScript or JavaScript...

PT-2026-52869

Summary Several LangChain components that resolve filesystem paths or expand search patterns do not consistently confine the resolved path to the intended root directory. Affected behaviors include: a file-search agent middleware that validates a starting directory but not the search pattern or t...

One-Click Microsoft 365 Copilot Flaw Could Have Let Attackers Steal Emails, Files, and MFA Codes

A single click on a trusted Microsoft link could have let an attacker pull emails, calendar details, and indexed files out of Microsoft 365 Copilot Enterprise Search. Researchers at Varonis Threat Labs chained three bugs into a one-click exfiltration path they call SearchLeak. Because the link...

SPARK: Security Knowledge Priming and Representation-Guided Knowledge Activation for LLM-Based Secure Code Generation

Large language models routinely generate code with exploitable security flaws. Prior literature attributes this limitation to a lack of security expertise, steering current defense mechanisms toward heavy fine-tuning or external knowledge retrieval, which introduces significant computational...

CVE-2026-42853

ApostropheCMS is an open-source Node.js content management system. Versions of the @apostrophecms/cli package up to and including 3.6.0 contain a command injection vulnerability in the apos create command. User-supplied input from the password prompt is embedded directly into a shell command...