Intelligence
Explorer.

lamp-cloud through 5.10.0 whitelists the path pattern /*/anno/** for anonymous access, allowing unauthenticated attackers to read the server's full JVM system property map. Attackers can send POST requests to /defGenProject/anno/getProperties to retrieve sensitive information including JVM classpath, filesystem paths, operating system details, and startup secrets.

pig before 4.1.0 contains an authentication bypass vulnerability in the /register/password endpoint where password verification results are discarded, allowing any value as the current password. Remote attackers can submit a username with an incorrect current password to overwrite any account credential including the admin account and gain full administrative control.

Semaphore UI through 2.19.12 exempts GET and HEAD requests from project resource permission checks in GetMustCanMiddleware. Attackers with guest or task_runner roles can read all project environments including plaintext secrets, credentials, and passwords via GET requests to the environment endpoint.

Jpom through 2.11.12 fails to validate workspace ownership when resolving repositoryId on the /build/branch-list endpoint, allowing authenticated users to access repositories from other workspaces. Attackers can submit repository identifiers from different workspaces to enumerate repository existence, determine repository type, and execute git ls-remote commands using other workspaces' stored credentials.

A flaw was found in gss-ntlmssp. A memory leak occurs in the NTLM target-info parser when a crafted NTLM CHALLENGE message contains duplicated string-valued AV_PAIR entries. The parser allocates memory for each string value but does not free the previous allocation when the same AV_PAIR type appears more than once, leaking the earlier allocation. A malicious or man-in-the-middle server can exploit this to cause gradual memory exhaustion on the client during NTLM authentication, leading to a denial of service.

An unauthenticated OS command injection vulnerability exists in the ping.php endpoint, allowing remote attackers to execute arbitrary commands on the underlying operating system and achieve remote code execution.

Polyaxon through 2.16.4 renders operation specification fields with an unsandboxed Jinja2 environment during server-side run preparation, allowing authenticated users to execute arbitrary code. Attackers can submit runs with Jinja2 payloads in queue, namespace, conditions, presets, or dependencies fields to execute operating system commands in the scheduler process context, exposing database credentials and service tokens.

pgweb through 0.17.0 leaves the POST /api/connect endpoint unguarded when connect-backend authorization is configured, allowing attackers to supply arbitrary database connection strings. Attackers can bypass the resource-to-database mapping by providing a custom session identifier and connection URL to access unauthorized databases and internal services.

KubeSphere through 4.1.3 contains a server-side request forgery vulnerability in the git credential verification endpoint that accepts unvalidated caller-supplied URLs without allowlist restrictions. Authenticated attackers can supply arbitrary URLs to reach internal services and exfiltrate basic-auth credentials from Secrets in any namespace by leveraging the endpoint's error response handling.

Steedos Platform through 3.0.15-beta.47 contains a reflected cross-site scripting vulnerability in the anonymous /api/page/render endpoint that fails to properly escape query parameters in inline script elements. Attackers can craft malicious links with script-terminating sequences in the schemaApi or data parameters to execute arbitrary JavaScript in victim sessions and steal X-Auth-Token credentials.

A flaw was found in GNOME Shell. When processing icons from a remote search provider via D-Bus, the system fails to validate the icon's declared dimensions against the actual data buffer size. A malicious or compromised remote search provider could exploit this by providing oversized icon dimensions, leading to an out-of-bounds read. This can cause the GNOME Shell process to crash, disrupting the user's session, and potentially disclose sensitive information from adjacent memory.

fast-uri is a dependency-free RFC 3986 URI parser for Node.js, used by Fastify and ajv, that added a mailto scheme parser in version 4.1.3. In versions 4.1.3 and 4.1.4, the mailto parser compares each query field name to the reserved names to, subject, and body while the name is still percent-encoded, and decodes it only when storing it as a generic header, so a percent-encoded spelling of a reserved field name is not recognized as that field at parse time but is re-emitted as the literal field name when the parsed URI is serialized. An application that validates, logs, or displays the recipient list from the first parse and then serializes the URI and sends it can silently gain an attacker-chosen recipient, and the subject and body fields can be smuggled across the same roundtrip. The issue is fixed in fast-uri 4.1.5, and users should upgrade to 4.1.5 or later. As a workaround, do not act on a mailto URI that fast-uri has re-serialized without first decoding and re-validating its recipient, subject, and body fields.

fast-uri is a dependency-free RFC 3986 URI parser for Node.js, used by Fastify and ajv. In versions before 2.4.7, from 3.0.0 through 3.1.7, and from 4.0.0 through 4.1.4, fast-uri folds the host to lowercase before it percent-decodes the host, so a percent-encoded uppercase octet such as %41 decodes to a literal A that is never folded. For a scheme-relative reference such as //host there is no scheme, so the host canonicalization that would normally repair this does not run, and parse, normalize, and equal then disagree on the same host. An application that makes a case-sensitive host decision on fast-uri output, for example a host allowlist or denylist that compares the parsed host or uses equal, can be steered past the check with a percent-encoded uppercase octet, and because hostnames are case-insensitive in DNS and HTTP the evading spelling still reaches the host the check meant to gate. The issue is fixed in fast-uri 2.4.7, 3.1.8, and 4.1.5, and users should upgrade to one of those versions or later. As a workaround, compare hosts case-insensitively by lowercasing the parsed host before any allowlist or denylist decision.

Converting crafted EUC_JISX0213 input to UCS-4 or the internal wide character encoding, for example with iconv, in the GNU C Library version 2.3 to 2.44 may result in the converter making no progress, causing the calling application to hang. Some EUC_JISX0213 sequences decode to two code points. If the output buffer has room for only the first one, the converter stores the second in the conversion state and returns E2BIG, but it never clears that pending character after emitting it on the next call. The converter then keeps emitting the pending character without consuming further input, so an application that retries the conversion loops forever. The input must be attacker controlled and the application must convert it with an output buffer small enough to split the two code points. Only the EUC_JISX0213 character set is affected, which is not commonly used. The related defect in SHIFT_JISX0213 converter is tracked separately as CVE-2026-77117.

Converting crafted SHIFT_JISX0213 input to UCS-4 or the internal wide character encoding, for example with iconv, in the GNU C Library version 2.3 to 2.44 may result in the converter making no progress, causing the calling application to hang. Some SHIFT_JISX0213 sequences decode to two code points. If the output buffer has room for only the first one, the converter stores the second in the conversion state and returns E2BIG, but it never clears that pending character after emitting it on the next call. The converter then keeps emitting the pending character without consuming further input, so an application that retries the conversion loops forever. The input must be attacker controlled and the application must convert it with an output buffer small enough to split the two code points. Only the SHIFT_JISX0213 character set is affected, which is not commonly used. The related defect in the EUC_JISX0213 converter is tracked separately as CVE-2026-80489.

PraisonAI is a multi-agent teams system. Prior to 0.1.6, praisonai_platform/services/auth_service.py falls back to the public dev-secret-change-me HS256 signing key when PLATFORM_JWT_SECRET is unset, while the startup and token-issuance guards are disabled because PLATFORM_ENV also defaults to dev. An unauthenticated attacker can sign a JWT containing an attacker-chosen sub value, and AuthService._verify_token() accepts it as an authenticated identity, enabling user or workspace-owner impersonation when a target identifier is known. This vulnerability is fixed in praisonai-platform 0.1.6.

PraisonAI is a multi-agent teams system. Prior to 0.1.6, praisonai_platform/services/auth_service.py assigns the public dev-secret-change-me value to JWT_SECRET when PLATFORM_JWT_SECRET is unset, and its production guard does not run when PLATFORM_ENV is also unset because that setting defaults to dev. A remote unauthenticated attacker can mint an HS256 token with an arbitrary sub and email, and the platform's AuthService._verify_token() and get_current_user dependency accept the forged identity for protected API routes. This vulnerability is fixed in praisonai-platform 0.1.6.

PraisonAI is a multi-agent teams system. Prior to 1.7.2, the codeMode tool in src/praisonai-ts/src/tools/builtins/code-mode.ts executes model-generated JavaScript with new Function() and with(sandbox), while a regular-expression blocklist can be bypassed with Function('return this')() to recover the global object and by constructing the child_process module name dynamically. An attacker who can influence the code argument can access host process capabilities, read or write files, obtain environment credentials, and execute operating-system commands with the PraisonAI process privileges. This issue is fixed in version 1.7.2.

PraisonAI is a multi-agent teams system. From 1.6.0 until 1.7.2, AgentOS in src/praisonai-ts/src/os/agentos.ts uses the 0.0.0.0 default from src/praisonai-ts/src/os/config.ts and registers GET /api/agents and POST /api/chat without authentication middleware. A remote caller who can reach the service can obtain agent names, roles, and instruction prefixes and can invoke a selected agent, potentially reaching its tools, memory, external APIs, credentials, and workflow state. An initial remediation was released in version 1.7.2.

PraisonAI is a multi-agent teams system. From 1.5.0 until 1.7.2, MCPServer.startHttp() in src/praisonai-ts/src/mcp/server.ts binds without a host restriction and forwards every HTTP POST request to handleRequest() without authentication or authorization. Any network client that can reach the port can call tools/list, tools/call, resources/read, or prompts/get, causing registered handlers to run with server-side credentials and process privileges or disclose registered data. An initial remediation was released in version 1.7.2.

PraisonAI is a multi-agent teams system. From 1.4.0 until 1.7.2, codeMode in src/praisonai-ts/src/tools/builtins/code-mode.ts executes untrusted JavaScript with new Function() inside with(sandbox) and relies on a small source-code blocklist plus shadowed process and require properties. Code can use ({}).constructor.constructor to recover the real Function constructor, obtain process and process.mainModule.require, and reach host filesystem and subprocess APIs despite the advertised sandbox. Attackers who control codeMode input can read secrets, modify files, execute commands, or exhaust the host process. This issue is fixed in version 1.7.2.

PraisonAI is a multi-agent teams system. From 1.4.0 until 1.7.2, createAgentLoop() in src/praisonai-ts/src/ai/agent-loop.ts passes executable tools to generateText() before invoking the onToolCall approval callback. Because the wrapped AI SDK executes tool handlers during generation, a callback that returns false records tool_rejected only after the denied tool has already produced side effects and populated toolResults. Applications using onToolCall as a human or policy approval boundary can therefore execute rejected file, command, API, or data-modifying operations. This issue is fixed in version 1.7.2.

PraisonAI is a multi-agent teams system. From 1.2.3 until 1.7.2, CommandValidator in src/praisonai-ts/src/cli/features/sandbox-executor.ts validates only the first whitespace-delimited executable against allowedCommands, then SandboxExecutor passes the complete command string to sh -c. A command beginning with an allowed executable can append a non-allowlisted command through shell metacharacters, causing arbitrary commands to run with the PraisonAI process privileges. This issue is fixed in version 1.7.2.

PraisonAI is a multi-agent teams system. From 1.2.3 until 1.7.2, SandboxExecutor network-isolated mode in src/praisonai-ts/src/cli/features/sandbox-executor.ts uses buildEnv() only to inject invalid http_proxy and https_proxy environment variables and does not establish an operating-system network boundary. Programs that ignore those proxy variables can open sockets directly, allowing supposedly isolated commands to reach localhost, internal services, cloud metadata, or external hosts and potentially exfiltrate data. An initial remediation was released in version 1.7.2.