The Yarbo cloud does not enforce per-device or per-user authorization. Any client possessing valid credentials, whether the shared hard-coded credentials or legitimate per-user credentials, can subscribe to wildcard topics covering all robots globally, and can publish to any robot's command topic using only the robot's serial number (disclosed in the telemetry stream). Even after removal of hard-coded credentials from the app, a single compromised credential could still provide fleet-wide access without per-device access controls.

Mem0 versions through 0.2.8, fixed in commit ae7f406, contain a missing authorization vulnerability in the self-hosted server component where the POST /configure endpoint modifies global LLM provider and embedder configuration but only verifies authentication via JWT or X-API-Key without validating the caller's role. Any authenticated user holding a distributed API key can redirect all LLM and embedder traffic to an attacker-controlled server, with the malicious configuration persisted to PostgreSQL and surviving server restarts to affect all users and API keys on the instance.

Unauthorized access to "/api/Token/gettoken" endpoint in EZD RP allows file manipulation.This issue affects EZD RP in versions before 20.19 (published on 22nd August 2024).

SQLBot is an intelligent data query system based on a large language model and RAG. Versions 1.5.0 and below contain a Stored Prompt Injection vulnerability that chains three flaws: a missing permission check on the Excel upload API allowing any authenticated user to upload malicious terminology, unsanitized storage of terminology descriptions containing dangerous payloads, and a lack of semantic fencing when injecting terminology into the LLM's system prompt. Together, these flaws allow an attacker to hijack the LLM's reasoning to generate malicious PostgreSQL commands (e.g., COPY ... TO PROGRAM), ultimately achieving Remote Code Execution on the database or application server with postgres user privileges. The issue is fixed in v1.6.0.

Coral Server is open collaboration infrastructure that enables communication, coordination, trust and payments for The Internet of Agents. Prior to 1.1.0, the SSE endpoint (/sse/v1/...) in Coral Server did not strongly validate that a connecting agent was a legitimate participant in the session. This could theoretically allow unauthorized message injection or observation. This vulnerability is fixed in 1.1.0.