Vulnerabilidades de HKUDS

CVE-2026-58173 Média · 6.5 CWE-22 30/06/2026

Vibe-Trading < 0.1.10 - Path Traversal via Persistent Memory Type

Vibe-Trading before 0.1.10 contains a path traversal vulnerability that allows attackers to write files outside the intended memory root directory by supplying a malicious memory_type value containing path traversal sequences through the remember tool. Attackers can manipulate the memory_type parameter in the persistent memory store to cause the application to write arbitrary Markdown files to unintended locations on the filesystem.

CVE-2026-58171 Média · 4.2 CWE-22 30/06/2026

Vibe-Trading < 0.1.10 - Path Traversal via Swarm Run Identifier

Vibe-Trading before 0.1.10 constructs the swarm run directory by joining a caller-supplied run identifier onto the runs base directory without validation in run_dir (agent/src/swarm/store.py). A crafted run identifier supplied through the MCP swarm tools causes the application to read arbitrary run.json files outside the runs directory and to overwrite existing run.json files at traversed locations.

CVE-2026-58170 Alta · 8.3 CWE-22 30/06/2026

Vibe-Trading < 0.1.10 - Path Traversal in Proposal Identifier Allows Forging Live Trading Mandates

Vibe-Trading before 0.1.10 builds the proposal file path by joining a caller-supplied proposal identifier onto the broker proposals directory without sanitization (agent/src/live/mandate/commit.py). A proposal identifier containing path traversal sequences causes the application to load an attacker-controlled JSON file as an authoritative live trading mandate. Combined with the file upload endpoint, an admitted caller can write a JSON file to a known location and traverse to it, and because the ceilings validation is skipped when ceilings are absent, the attacker fully controls the committed mandate.

CVE-2026-58169 Alta · 7.5 CWE-346 30/06/2026

Vibe-Trading < 0.1.10 - Loopback Trust and Missing Host Validation Enable DNS-Rebinding Authentication Bypass and Remote Code Execution

Vibe-Trading before 0.1.10 contains a DNS rebinding authentication bypass vulnerability that allows remote attackers to bypass bearer-token authentication by exploiting the server's trust of TCP peer addresses for loopback clients combined with missing Host header validation while binding to 0.0.0.0 with credentialed CORS. Attackers can craft a malicious DNS rebinding page to issue authenticated requests to the local API server, reach the shell execution endpoint with a bash-enabled preset, and achieve remote code execution as the API process user while also overwriting LLM and data-source settings to exfiltrate credentials.

CVE-2026-58168 Alta · 8.8 CWE-862 30/06/2026

DeepTutor < 1.4.10 - Insecure Default Grants Unrestricted MCP Tool Access to Non-Admin Users

DeepTutor before version 1.4.10 contains an authorization bypass vulnerability that allows low-privilege users to invoke unrestricted MCP tools due to the allowed_mcp_tools function returning None instead of a denied result when mcp_tools is omitted from a user's grant in deeptutor/multi_user/tool_access.py. Attackers or prompt-injected content acting within a user session can enumerate and invoke any configured MCP tool, including filesystem, shell, and browser servers, gaining unauthorized access to sensitive deployment resources.

CVE-2026-56696 Média · 5.4 CWE-862 23/06/2026

OpenHarness - Prompt Injection via /issue and /pr_comments Slash Commands

OpenHarness /issue and /pr_comments slash commands lack remote_invocable=False protection, allowing remote channel senders to write attacker-controlled Markdown into project context files. Admitted remote attackers can inject malicious content into .openharness/issue.md and .openharness/pr_comments.md files, which are subsequently injected into runtime system prompts, persistently influencing local agent behavior.

CVE-2026-56695 Média · 6.5 CWE-862 23/06/2026

OpenHarness - Cross-Session Disclosure via /resume and /summary Commands

OpenHarness ohmo gateway /resume and /summary slash commands default remote_invocable to True, allowing admitted remote senders to enumerate and load arbitrary session snapshots by ID. Attackers can exploit this to access victim snapshots containing private prompts, credentials, tool output, and file paths via shared gateway channels.

CVE-2026-48716 Alta · 8.7 CWE-22 18/06/2026

nanobot: Path traversal via unsanitized WhatsApp document fileName enables arbitrary file write

nanobot is a personal AI assistant. In versions 0.1.5.post3 and prior, the WhatsApp bridge in bridge/src/whatsapp.ts constructs a filesystem path using the fileName field from an incoming WhatsApp document message without sanitization. The WhatsApp bridge downloads media attachments and writes them to disk using a filename derived from the sender's message via documentMessage.fileName, which is concatenated with a prefix and its raw value is passed directly to path.join(mediaDir, outFilename). Node.js path.join resolves .. components, allowing an attacker to escape the intended media/ directory by sending a document with a crafted fileName such as ../../../.ssh/authorized_keys. Because the attacker also controls the file content (the downloaded buffer), this is a write-anywhere primitive — both path and content are attacker-controlled. A fix for this issue is planned for version 0.1.5.post4.

CVE-2026-12203 Média · 5.3 CWE-200 14/06/2026

HKUDS AI-Trader Research Export agents.csv information disclosure

A vulnerability was found in HKUDS AI-Trader up to 74caf996f78dcc0c657df8365c8544678a16e215. This affects an unknown part of the file /api/research/agents.csv of the component Research Export. Performing a manipulation results in information disclosure. Remote exploitation of the attack is possible. The exploit has been made public and could be used. This product follows a rolling release approach for continuous delivery, so version details for affected or updated releases are not provided. The patch is named 91a31aac1b0f4dbc6b8bef9f6eff0b7912e0bc65. Applying a patch is the recommended action to fix this issue. The vendor confirms: "Research export endpoints now require an authenticated agent with the research_exports capability".

CVE-2026-49140 Média · 4.3 CWE-770 01/06/2026

Nanobot < 0.2.1 Denial of Service via Matrix Media Download Handler

Nanobot prior to version 0.2.1 contains a denial of service vulnerability in the Matrix channel media download handler that allows authenticated room members to exhaust process memory and bandwidth by sending media events with missing or invalid size metadata. Attackers can send multiple concurrent Matrix media events with omitted or invalid declared sizes to trigger simultaneous large media downloads that fully materialize response bodies before post-download rejection, consuming process resources until service degradation occurs.

CVE-2026-49139 Alta · 7.0 CWE-918 01/06/2026

Nanobot < 0.2.1 SSRF via Microsoft Teams Channel serviceUrl Poisoning

Nanobot prior to version 0.2.1 contains a server-side request forgery vulnerability in the Microsoft Teams channel handler that allows remote attackers to exfiltrate Bot Framework bearer tokens by supplying a forged activity with an attacker-controlled serviceUrl value. Attackers can poison the stored conversation reference by sending a crafted inbound activity to the Teams webhook, causing subsequent bot replies to transmit token-bearing Authorization header requests to an attacker-controlled host.

CVE-2026-49138 Média · 5.0 CWE-918 01/06/2026

Nanobot < 0.2.1 SSRF via web_fetch Tool Redirect Following

Nanobot prior to version 0.2.1 contains a server-side request forgery vulnerability in the web_fetch tool that allows remote attackers to reach internal or private network hosts by supplying a URL that redirects to a loopback or private address via a 3xx Location header. Attackers can exploit the automatic HTTP redirect following behavior in the httpx library to bypass initial URL validation and cause the runtime to send outbound requests to internal hosts before final resolved URL validation is applied.

CVE-2026-32847 Alta · 7.5 CWE-22 28/05/2026

DeepCode 1.2.0 Path Traversal via SPA Catch-All Route in main.py

DeepCode through commit c991dc2 contains a path traversal vulnerability in the SPA catch-all route in new_ui/backend/main.py that allows unauthenticated attackers to read arbitrary files by supplying percent-encoded path segments to the GET /{full_path:path} endpoint. Attackers can bypass Starlette's path normalization by encoding slashes as %2F and dots as %2E%2E, causing the joined path to traverse outside FRONTEND_DIST and exposing sensitive files such as SSH private keys, TLS certificates, and application secrets with a single HTTP request.

CVE-2026-7551 Alta · 8.8 CWE-78 30/04/2026

HKUDS OpenHarness Remote Command Execution via /bridge Slash Command

HKUDS OpenHarness contains a remote code execution vulnerability in the /bridge slash command that allows remote senders accepted by configuration to execute arbitrary operating system commands. Attackers can invoke the /bridge spawn command with attacker-controlled command text that is forwarded to the bridge session manager and executed through the shared shell subprocess helper, allowing them to spawn shell sessions as the OpenHarness process user and access local files, credentials, workspace state, and repository contents.

CVE-2026-6823 Alta · 8.2 CWE-276 21/04/2026

HKUDS OpenHarness Insecure Default Remote Channel Allowlist

HKUDS OpenHarness prior to PR #147 remediation contains an insecure default configuration vulnerability where remote channels inherit allow_from = ["*"] permitting arbitrary remote senders to pass admission checks. Attackers who can reach the configured channel can bypass access controls and reach host-backed agent runtimes, potentially leading to unauthorized file disclosure and read access through default-enabled read-only tools.

CVE-2026-6819 Alta · 8.8 CWE-276 21/04/2026

HKUDS OpenHarness Plugin Management Command Exposure

HKUDS OpenHarness prior to PR #156 remediation exposes plugin lifecycle commands including /plugin install, /plugin enable, /plugin disable, and /reload-plugins to remote senders by default. Attackers who gain access through the channel layer can remotely manage plugin trust and activation state, enabling unauthorized plugin installation and activation on the system.

CVE-2026-6729 Alta · 7.6 CWE-287 20/04/2026

HKUDS OpenHarness Session Key Collision Privilege Escalation

HKUDS OpenHarness prior to PR #159 remediation contains a session key derivation vulnerability that allows authenticated participants in shared chats or threads to hijack other users' sessions by exploiting a shared ohmo session key that lacks sender identity verification. Attackers can reuse another user's conversation state and replace or interrupt their active tasks by colliding into the same session boundary through the shared chat or thread scope.

CVE-2026-40516 Média · 6.3 CWE-918 17/04/2026

OpenHarness SSRF via web_fetch and web_search

OpenHarness before commit bd4df81 contains a server-side request forgery vulnerability in the web_fetch and web_search tools that allows attackers to access private and localhost HTTP services by manipulating tool parameters without proper validation of target addresses. Attackers can influence an agent session to invoke these tools against loopback, RFC1918, link-local, or other non-public addresses to read response bodies from local development services, cloud metadata endpoints, admin panels, or other private HTTP services reachable from the victim host.

CVE-2026-40515 Média · 5.5 CWE-863 17/04/2026

OpenHarness Permission Bypass via grep and glob root argument

OpenHarness before commit bd4df81 contains a permission bypass vulnerability that allows attackers to read sensitive files by exploiting incomplete path normalization in the permission checker. Attackers can invoke the built-in grep and glob tools with sensitive root directories that are not properly evaluated against configured path rules, allowing disclosure of sensitive local file content, key material, configuration files, or directory contents despite configured path restrictions.

CVE-2026-40503 Média · 6.5 CWE-22 15/04/2026

OpenHarness Path Traversal Information Disclosure via /memory show

OpenHarness prior to commit dd1d235 contains a path traversal vulnerability that allows remote gateway users with chat access to read arbitrary files by supplying path traversal sequences to the /memory show slash command. Attackers can manipulate the path input parameter to escape the project memory directory and access sensitive files accessible to the OpenHarness process without filesystem containment validation.

CVE-2026-40502 Alta · 8.8 CWE-862 15/04/2026

OpenHarness Remote Administrative Command Injection via Gateway Handler

OpenHarness prior to commit dd1d235 contains a command injection vulnerability that allows remote gateway users with chat access to invoke sensitive administrative commands by exploiting insufficient distinction between local-only and remote-safe commands in the gateway handler. Attackers can execute administrative commands such as /permissions full_auto through remote chat sessions to change permission modes of a running OpenHarness instance without operator authorization.

CVE-2026-35589 Crítica · 9.3 CWE-1385 14/04/2026

nanobot: Cross-Site WebSocket Hijacking in WhatsApp Bridge (CVE-2026-2577 Fix Update)

nanobot is a personal AI assistant. Versions prior to 0.1.5 contain a Cross-Site WebSocket Hijacking (CSWSH) vulnerability exists in the bridge's WebSocket server in bridge/src/server.ts, resulting from an incomplete remediation of CVE-2026-2577. The original fix changed the binding from 0.0.0.0 to 127.0.0.1 and added an optional BRIDGE_TOKEN parameter, but token authentication is disabled by default and the server does not validate the Origin header during the WebSocket handshake. Because browsers do not enforce the Same-Origin Policy on WebSockets unless the server explicitly denies cross-origin connections, any website visited by a user running the bridge can establish a WebSocket connection to ws://127.0.0.1:3001/ and gain full access to the bridge API. This allows an attacker to hijack the WhatsApp session, read incoming messages, steal authentication QR codes, and send messages on behalf of the user. This issue has bee fixed in version 0.1.5.

CVE-2026-39413 Média · 6.5 CWE-347 08/04/2026

LightRAG has a JWT Algorithm Confusion Vulnerability in LightRAG API

LightRAG provides simple and fast retrieval-augmented generation. Prior to 1.4.14, the LightRAG API is vulnerable to a JWT algorithm confusion attack where an attacker can forge tokens by specifying 'alg': 'none' in the JWT header. Since the jwt.decode() call does not explicitly deny the 'none' algorithm, a crafted token without a signature will be accepted as valid, leading to unauthorized access. This vulnerability is fixed in 1.4.14.

CVE-2026-22682 Alta · 7.1 CWE-863 07/04/2026

OpenHarness Improper Access Control via File Tools

OpenHarness prior to commit 166fcfe contains an improper access control vulnerability in built-in file tools due to inconsistent parameter handling in permission enforcement, allowing attackers who can influence agent tool execution to read arbitrary local files outside the intended repository scope. Attackers can exploit the path parameter not being passed to the PermissionChecker in read_file, write_file, edit_file, and notebook_edit tools to bypass deny rules and access sensitive files such as configuration files, credentials, and SSH material, or create and overwrite files in restricted host paths in full_auto mode.

CVE-2026-33654 Crítica · 9.8 CWE-94 27/03/2026

Zero-Click Indirect Prompt Injection and Authentication Bypass via Email Polling

nanobot is a personal AI assistant. Prior to version 0.1.6, an indirect prompt injection vulnerability exists in the email channel processing module (`nanobot/channels/email.py`), allowing a remote, unauthenticated attacker to execute arbitrary LLM instructions (and subsequently, system tools) without any interaction from the bot owner. By sending an email containing malicious prompts to the bot's monitored email address, the bot automatically polls, ingests, and processes the email content as highly trusted input, fully bypassing channel isolation and resulting in a stealthy, zero-click attack. Version 0.1.6 patches the issue.