Vulnerabilities
All
My stack
Known exploited
Public exploit
Critical (9+)
Likely exploited (EPSS ≥ 50%)
Changed, 24 h
New, 24 h
Sort:
risk
modified
published
CVSS
EPSS
| CVE | CVSS | EPSS | Vendor, product | Description | Modified |
|---|---|---|---|---|---|
| CVE-2026-47830 | 8.5 high | 0.00% | no CPE yet | Incorrect Permission Assignment in BOSH.Utils.psm1 in BOSH-Ecosystem bosh-windows-stemcell-builder allows low-privilege authenticated users to overwrite C:\bosh\service_wrapper.exe or C:\bosh\bosh-agent.exe and gain NT AUTHORITY\SYSTEM on the next service restart or reboot. This can lead to full host control. Affected versions: bosh-windows-stemcell-builder versions prior to v2019.98. | Sun 25 Oct |
| CVE-2026-67424 | 8.5 high | 0.00% | no CPE yet | Flyto2 Core is an execution kernel for automation and AI-agent workflows. Prior to 2.26.7, the HTTP modules http.get, http.request, and http.batch in src/core/modules/atomic/http/get.py, src/core/modules/atomic/http/request.py, and src/core/modules/atomic/http/batch.py validate only the initial URL, then follow redirects with allow_redirects=True and without per-hop Location revalidation, allowing a public URL to redirect into internal address space and return the internal response body. This issue is fixed in version 2.26.7. | Sat 24 Oct |
| CVE-2026-67428 | 8.5 high | 0.00% | no CPE yet | Flyto2 Core is an execution kernel for automation and AI-agent workflows. Prior to 2.26.7, HTTP-emitting modules including src/core/modules/third_party/developer/http/requests.py, core.api.http_get, core.api.http_post, graphql.query, graphql.mutation, monitor.http_check, communication.slack_send, notification.discord.send_message, notification.slack.send_message, notification.teams.send_message, ai.vision_analyze, verify.visual_diff, browser.proxy_rotate, and the agent and llm inline base_url branch fetch caller-controlled URLs without validate_url_with_env_config, allowing SSRF to internal or metadata endpoints. This issue is fixed in version 2.26.7. | Sat 24 Oct |
| CVE-2026-91093 | 8.5 high | 0.00% | no CPE yet | A potential security vulnerability has been identified in HP One Agent for certain HP PC products, which might allow for escalation of privilege and/or arbitrary code execution. HP is releasing software updates to mitigate the potential vulnerability. | Sat 24 Oct |
| CVE-2026-75858 | 8.5 high | 0.00% | no CPE yet | CodeWhale (packages codewhale / codewhale-tui) versions >= 0.8.41 and < 0.8.64 contain a remote code execution vulnerability in the rlm_eval tool. The tool's approval_requirement() returns ApprovalRequirement::Auto, which the engine treats as 'never prompt,' causing arbitrary model-supplied Python code to run in a python3 interpreter without consulting the user's configured --approval-policy and without any approval prompt or audit step. An attacker can induce the agent to execute arbitrary code via prompt injection in untrusted content the agent reads (a web page, fetched URL, repository file, or MCP tool result); the companion rlm_open tool can stage such content. Code runs on the user's machine at the user's privilege level. Fixed in 0.8.64. | 22 Oct |
| CVE-2026-50218 | 8.5 high | 0.00% | no CPE yet | Akamai GPA (Guardicore Platform Agent) client on all platforms before 7.4 improperly validates JWT tokens used for seeding. By convincing a user to interact with a malicious link, an attacker can use a spoofed JWT token to inject arbitrary parameters into the client software, resulting in unauthorized reconfiguration and full compromise of the application's functionality. This attack can also be targeted against an already seeded client if the attacker knows the URL of the IDP that the client used for enrollment. | 19 Oct |
| CVE-2025-34352 | 8.5 high | 0.00% | no CPE yet | JumpCloud Remote Assist for Windows versions prior to 0.317.0 include an uninstaller that is invoked by the JumpCloud Windows Agent as NT AUTHORITY\SYSTEM during agent uninstall or update operations. The Remote Assist uninstaller performs privileged create, write, execute, and delete actions on predictable files inside a user-writable %TEMP% subdirectory without validating that the directory is trusted or resetting its ACLs when it already exists. A local, low-privileged attacker can pre-create the directory with weak permissions and leverage mount-point or symbolic-link redirection to (a) coerce arbitrary file writes to protected locations, leading to denial of service (e.g., by overwriting sensitive system files), or (b) win a race to redirect DeleteFileW() to attacker-chosen targets, enabling arbitrary file or folder deletion and local privilege escalation to SYSTEM. This issue is fixed in JumpCloud Remote Assist 0.317.0 and affects Windows systems where Remote Assist is installed and managed through the Agent lifecycle. | 18 Oct |
| CVE-2026-107322 | 8.5 high | 0.00% | no CPE yet | An incomplete list of disallowed inputs in Amazon Agent Plugins for AWS databases-on-aws plugin before 1.7.1 might allow a remote unauthenticated actor to execute arbitrary operating system commands on the host running the helper via a crafted database command value introduced in the agent context. To remediate this issue, users should upgrade to databases-on-aws plugin version 1.7.1 or later and verify that the updated plugin is active in each environment where it is used. | 8 Oct |
| CVE-2026-107347 | 8.4 high | — | hashicorp, nomad, nomad enterprise | HashiCorp Nomad and Nomad Enterprise are vulnerable to symlink following during dispatch payload and executor log writes that may allow a malicious workload to modify files outside its allocation directory. When the client agent runs as root, this could lead to code execution as root on the client host. This vulnerability (CVE-2026-107347) is fixed in Nomad 2.1.1 and Nomad Enterprise 2.1.0, 1.11.12, and 1.10.18. | 7 h ago |
| CVE-2026-65151 | 8.4 high | 0.00% | no CPE yet | NVIDIA NeMo Agent Toolkit contains vulnerability where an attacker could cause a SQL injection issue. A successful exploit of this vulnerability might lead to data tampering, denial of service and information disclosure. | Tue 27 Oct |
| CVE-2026-22682 | 8.4 high | 0.00% | no CPE yet | 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. | Mon 26 Oct |
| CVE-2026-40287 | 8.4 high | 0.00% | praison, praisonai, praisonaiagents | PraisonAI is a multi-agent teams system. Versions 4.5.138 and below are vulnerable to arbitrary code execution through automatic, unsanitized import of a tools.py file from the current working directory. Components including call.py (import_tools_from_file()), tool_resolver.py (_load_local_tools()), and CLI tool-loading paths blindly import ./tools.py at startup without any validation, sandboxing, or user confirmation. An attacker who can place a malicious tools.py in the directory where PraisonAI is launched (such as through a shared project, cloned repository, or writable workspace) achieves immediate arbitrary Python code execution in the host environment. This compromises the full PraisonAI process, the host system, and any connected data or credentials. This issue has been fixed in version 4.5.139. | Mon 26 Oct |
| CVE-2026-41433 | 8.4 high | 0.00% | opentelemetry, opentelemetry_ebpf_instrumentation | OpenTelemetry eBPF Instrumentation provides eBPF instrumentation based on the OpenTelemetry standard. From 0.4.0 to before 0.8.0, a flaw in the Java agent injection path allows a local attacker controlling a Java workload to overwrite arbitrary host files when Java injection is enabled and OBI is running with elevated privileges. The injector trusted TMPDIR from the target process and used unsafe file creation semantics, enabling both filesystem boundary escape and symlink-based file clobbering. This vulnerability is fixed in 0.8.0. | Mon 26 Oct |
| CVE-2026-44334 | 8.4 high | 0.00% | praison, praisonai | PraisonAI is a multi-agent teams system. From version 4.5.139 to before version 4.6.32, CVE-2026-40287's fix gated tools.py auto-import behind PRAISONAI_ALLOW_LOCAL_TOOLS=true in two files (tool_resolver.py, api/call.py). A third import sink in praisonai/templates/tool_override.py was missed and remains unguarded. It is reached by the recipe runner on every recipe execution and is remotely triggerable through POST /v1/recipes/run with a recipe value pointing at any local absolute path or any GitHub repo (because SecurityConfig.allow_any_github defaults to True). The attacker drops a tools.py next to TEMPLATE.yaml; the server exec_module()s it. No auth required by default, no environment opt-in required. This issue has been patched in version 4.6.32. | Mon 26 Oct |
| CVE-2026-43990 | 8.4 high | 0.00% | no CPE yet | JunoClaw is an agentic AI platform built on Juno Network. Prior to 0.x.y-security-1, plugin-shell's run_command wrapped every agent-supplied command in 'sh -c' / 'cmd /C' and passed the full argument string to the shell's parser, allowing shell metacharacters in agent-supplied arguments to be interpreted as command syntax. This vulnerability is fixed in 0.x.y-security-1. | Mon 26 Oct |
| CVE-2026-54320 | 8.4 high | 0.00% | no CPE yet | Daytona is a secure and elastic infrastructure runtime for AI-generated code execution and agent workflows. Prior to 0.184.0, organization invitations could be accepted (and declined) by a user whose email matched the invitation but had not been verified. Daytona authenticates users via OIDC and matches an invitation's target email against the email in the caller's token, but the invitation accept and decline paths did not require that email to be verified, unlike organization creation, which already enforced verification. On identity providers that allow self-service signup and issue a session before the email is verified, an actor could register an address matching a pending invitation, leave it unverified, and accept the invitation, joining the target organization with the role the invitation carried (up to Owner). This vulnerability is fixed in 0.184.0. | Sun 25 Oct |
| CVE-2026-66065 | 8.4 high | 0.00% | no CPE yet | Ouroboros is a local-first runtime for AI coding agents that records their actions and applies user-defined policies to constrain behavior. Versions prior to 0.42.1 have an incomplete denylist. Several execution-routing keys of the same RCE class were omitted, so a malicious cloned repo can still reach arbitrary command execution by shipping a .env (auto-loaded at import, with no review step). The CVE-2026-47211 fix added _UNTRUSTED_ENV_DENYLIST to stop an untrusted project-directory .env from redirecting execution, but it did not account for all keys. The backend config-home and MCP/plugin roots bypass the approval gate by pointing the nested agent, MCP servers, and plugin roster at attacker config. Other variables re-enable blocked local transports, replace sub-agent prompts, switch backends, and lower tool approval classes, further weakening the approval gate. This issue has been fixed in version 0.42.1. | 23 Oct |
| CVE-2026-55786 | 8.4 high | 0.00% | no CPE yet | Flyto2 Core is an execution kernel for automation and AI-agent workflows. From 2.26.2 until 2.26.4, the HTTP MCP POST /mcp route accepts unauthenticated JSON-RPC tools/call requests and forwards execute_module arguments to the module registry without the require_auth dependency used by the REST execution routes. An unauthenticated caller can select sandbox.execute_shell and supply its command parameter, which reaches asyncio.create_subprocess_shell with the privileges of the flyto-core server process. The endpoint is reachable locally under the default loopback binding and becomes network reachable when an operator binds the server to a non-loopback interface, allowing arbitrary OS command execution and potential full host compromise. This issue is fixed in version 2.26.4. | 16 Oct |
| CVE-2026-72343 | 8.4 high | 0.00% | no CPE yet | In the Linux kernel, the following vulnerability has been resolved: net/mlx5e: Fix HV VHCA stats zero-sized buffer allocation mlx5e_hv_vhca_stats_create() is called from mlx5e_nic_enable(), before mlx5e_open(). At that point priv->stats_nch is still zero, because it is only ever incremented in mlx5e_channel_stats_alloc(), which is reached only from mlx5e_open_channel(). mlx5e_hv_vhca_stats_buf_size() therefore returns 0, and kvzalloc(0, GFP_KERNEL) returns ZERO_SIZE_PTR ((void *)16) rather than NULL. The "if (!buf)" guard does not catch this, and mlx5e_hv_vhca_stats_create() completes "successfully" with priv->stats_agent.buf set to ZERO_SIZE_PTR. Once channels are opened (priv->stats_nch > 0) and the hypervisor enables stats reporting, mlx5e_hv_vhca_stats_work() recomputes buf_len using the new non-zero stats_nch and calls memset(buf, 0, buf_len) on ZERO_SIZE_PTR, faulting at address 0x10. Allocate the buffer based on priv->max_nch, which is set in mlx5e_priv_init() and is the upper bound on stats_nch: - Add a separate helper mlx5e_hv_vhca_stats_buf_max_size() that returns sizeof(per_ring_stats) * max(max_nch, stats_nch), and use it for the kvzalloc() in mlx5e_hv_vhca_stats_create(). - Keep mlx5e_hv_vhca_stats_buf_size() (which returns based on stats_nch) for the worker's active payload size, so the wire format (block->rings = stats_nch) and the amount of data filled by mlx5e_hv_vhca_fill_stats() are unchanged. The max(max_nch, stats_nch) guard handles the rare case where mlx5e_attach_netdev() recomputes max_nch downward across a detach/resume cycle while priv->stats_nch persists (mlx5e_detach_netdev does not call mlx5e_priv_cleanup, so stats_nch is only reset when the netdev is destroyed). Without the guard, the worker could compute buf_len from stats_nch and overrun the smaller buffer allocated based on the reduced max_nch. Allocating a non-zero buffer also makes the kvzalloc() failure path in mlx5e_hv_vhca_stats_create() reachable for the first time: it returns early without (re)creating the agent. Clear priv->stats_agent.{agent,buf} in mlx5e_hv_vhca_stats_destroy() after freeing them, so that if a later create() bails out on this path, a subsequent teardown does not double-free the stale agent/buffer left from a previous enable/disable cycle. This mirrors the existing mlx5e pattern of preallocating arrays of size max_nch (e.g. priv->channel_stats) and lazily populating entries up to stats_nch on demand. | 16 Oct |
| CVE-2026-72342 | 8.4 high | 0.00% | no CPE yet | In the Linux kernel, the following vulnerability has been resolved: net/mlx5e: Fix HV VHCA stats agent registration race mlx5e_hv_vhca_stats_create() registers the stats agent through mlx5_hv_vhca_agent_create(). The helper publishes the agent in hv_vhca->agents[type] under agents_lock and immediately schedules an asynchronous control invalidation on the HV VHCA workqueue before returning to mlx5e. The asynchronous invalidation invokes the control agent's invalidate callback, which reads the hypervisor control block and forwards the command to mlx5e_hv_vhca_stats_control(). That callback may either: - call cancel_delayed_work_sync(&priv->stats_agent.work), or - call queue_delayed_work(priv->wq, &sagent->work, sagent->delay). However, the delayed_work and priv->stats_agent.agent are only initialized after mlx5_hv_vhca_agent_create() returns to mlx5e: agent = mlx5_hv_vhca_agent_create(...); /* publish + invalidate */ ... priv->stats_agent.agent = agent; /* too late */ INIT_DELAYED_WORK(&priv->stats_agent.work, ...); /* too late */ If the asynchronous control path runs before the two assignments above, it can: - Operate on an uninitialized delayed_work whose timer.function is NULL. queue_delayed_work() calls add_timer() unconditionally, so when the timer expires the timer softirq invokes a NULL function pointer. - Re-initialize the timer later through INIT_DELAYED_WORK() while the timer is already enqueued in the timer wheel, corrupting the hlist (entry.pprev cleared while the previous bucket node still points at this entry). - When the worker eventually runs, mlx5e_hv_vhca_stats_work() reads sagent->agent (NULL) and dereferences it inside mlx5_hv_vhca_agent_write(). Fix this by: - Initializing priv->stats_agent.work before invoking mlx5_hv_vhca_agent_create(), so the work is always in a valid state when the control callback observes it. - Adding a struct mlx5_hv_vhca_agent **ctx_update out-parameter to mlx5_hv_vhca_agent_create(). The helper writes the agent pointer to *ctx_update before publishing into hv_vhca->agents[] and triggering the agents_update flow, so any callback subsequently invoked from that flow already sees a valid priv->stats_agent.agent. This avoids having the control callback participate in agent initialization. While at it, access priv->stats_agent.agent with READ_ONCE()/WRITE_ONCE() for the cross-CPU access with the worker, and clear priv->stats_agent.buf on the agent_create() failure path. | 16 Oct |
| CVE-2026-118687 | 8.3 high | 0.00% | no CPE yet | Coder before 2.29.21, 2.30.0 before 2.34.12, 2.35.0 before 2.35.9, 2.36.0 before 2.36.7, and 2.37.0 before 2.37.4 does not validate the audience claim of GCE instance-identity JWTs at the POST /api/v2/workspaceagents/google-instance-identity endpoint. Attackers who obtain an unexpired format=full identity token minted for another relying party can replay it to obtain that workspace agent's session token, which remains valid until the workspace's next build. | Wed 28 Oct |
| CVE-2026-116794 | 8.3 high | 0.00% | no CPE yet | russh is a Rust SSH client and server library that includes an SSH agent protocol implementation. From 0.50.0-beta.12 until 0.65.0, the agent server's Connection::try_sign path in russh/src/keys/agent/server.rs recognizes SSH_AGENT_CONSTRAIN_CONFIRM but returns a hard-coded successful result instead of calling Agent::confirm. In applications that override Agent::confirm to require user approval or another policy check, a process that can reach the agent socket, including the remote side of a forwarded agent, can use a key added through ssh-add -c without that check and can authenticate or sign data as the user while the key remains loaded. The separate Agent::confirm_request hook still runs for every signing request, so checks implemented there remain effective, and implementations that retain the default always-successful Agent::confirm do not lose an approval check through this regression. This issue is fixed in version 0.65.0. | Wed 28 Oct |
| CVE-2026-6823 | 8.3 high | 0.00% | hkuds, openharness | 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. | Mon 26 Oct |
| CVE-2026-6442 | 8.3 high | 0.00% | no CPE yet | Improper validation of bash commands in Snowflake Cortex Code CLI versions prior to 1.0.25 allowed subsequent commands to execute outside the sandbox. An attacker could exploit this by embedding specially crafted commands in untrusted content, such as a malicious repository, causing the CLI agent to execute arbitrary code on the local device without user consent. Exploitation is non-deterministic and model-dependent. The fix is automatically applied upon relaunch with no user action required. | Sun 25 Oct |
| CVE-2026-49471 | 8.3 high | 0.00% | oraios-ai, serena | Serena is a powerful MCP toolkit for coding that provides semantic retrieval and editing capabilities. Prior to v1.5.2, Serena's built-in web dashboard exposes an unauthenticated Flask API on a fixed, predictable port, with no authentication, no CSRF protection, and no Host header validation. A DNS rebinding attack allows a malicious webpage to reach this API from any browser and write arbitrary content to the agent's persistent memory store, which the agent reads and acts on autonomously. Combined with execute_shell_command using shell=True, this creates a remote code execution chain requiring only that the victim visit a malicious webpage while Serena is running. This issue is fixed in version v1.5.2. | Sun 25 Oct |
| CVE-2026-47415 | 8.3 high | 0.00% | no CPE yet | PraisonAI Platform is the platform layer for the PraisonAI multi-agent teams system. Versions prior to 0.1.4 have an Insecure Direct Object Reference. The issue CRUD endpoints (`GET / PATCH / DELETE /workspaces/{workspace_id}/issues/{issue_id}`) gate access on `require_workspace_member(workspace_id)` only, then resolve `issue_id` through `IssueService.get(issue_id)` which is a primary-key lookup with no workspace constraint. A user who is a member of any workspace `W1` can read, modify, or delete issues that belong to a different workspace `W2`. PraisonAI Platform version 0.1.4 patches the issue. | Sat 24 Oct |
| CVE-2026-47419 | 8.3 high | 0.00% | no CPE yet | PraisonAI Platform is the platform layer for the PraisonAI multi-agent teams system. Versions prior to 0.1.4 have an* Insecure Direct Object Reference. The agent CRUD endpoints (`GET / PATCH / DELETE /workspaces/{workspace_id}/agents/{agent_id}`) gate access on `require_workspace_member(workspace_id)` only, then resolve `agent_id` through `AgentService.get(agent_id)` which is a primary-key lookup with no workspace constraint. A user who is a member of any workspace `W1` can read, modify, or delete agents that belong to a different workspace `W2` by guessing or harvesting an agent UUID and calling `…/workspaces/W1/agents/<W2-agent-id>`. PraisonAI Platform version 0.1.4 patches the issue. | Sat 24 Oct |
| CVE-2026-56835 | 8.3 high | 0.00% | no CPE yet | PraisonAI is a multi-agent teams system. From version 3.11.0 to before version 4.6.59, PraisonAI's Slack bot applies its configured allowed_users, allowed_channels, and unknown-user pairing policy in the normal Slack message event handler, but not in the adjacent Slack app_mention event handler. A Slack workspace user who can mention the bot in a channel where the Slack app is present can trigger the configured PraisonAI agent even when: the sender is not in BotConfig.allowed_users; the channel is not in BotConfig.allowed_channels; unknown_user_policy="deny" is configured; and the same event content is correctly dropped by the normal message handler. This issue has been patched in version 4.6.59. | 23 Oct |
| CVE-2026-112308 | 8.3 high | 0.00% | no CPE yet | crewai-tools through 1.15.28 contains a server-side request forgery vulnerability in SeleniumScrapingTool that allows attackers to fetch internal resources because it never applies the validate_url private-address check. Attackers influencing agent tool input through prompt injection can direct headless Chrome to loopback, RFC1918, or cloud metadata addresses to disclose internal data and cloud credentials. | 19 Oct |
| CVE-2026-110769 | 8.3 high | 0.00% | no CPE yet | holaOS through 2026.822.1 contains a missing authorization vulnerability in channel-gateway connectors that allows any messaging-platform user to submit agent tasks because the #isAllowed check accepts all senders when the default allowFrom list is empty. Attackers who can message the operator's Telegram, Slack, Discord, or other bot can have ChannelIngress.handle pass turns to the workspace agent, running configured tools on the operator's machine. | 15 Oct |
| CVE-2026-76072 | 8.3 high | 0.00% | no CPE yet | The Continue CLI applies an incomplete denylist as its only barrier to destructive shell commands when running unattended. In headless mode and auto mode the default policy in extensions/cli/src/permissions/defaultPolicies.ts grants the Bash tool the allow permission, and permissionChecker.ts hard-blocks a command only when the terminal-security evaluator returns a disabled verdict, so isCriticalCommand in packages/terminal-security/src/evaluateTerminalCommandSecurity.ts is the sole control. Its dangerous-path test matches only /, /*, ~, ~/*, /usr, /etc, /bin and /sbin and their prefixes, so a recursive forced removal of /home, /root, /var, /opt or /srv is not disabled. The command line is parsed with shell-quote, which reduces $HOME to an empty token, so rm -rf $HOME also fails the dangerous-path test while the shell re-expands the variable when the command is spawned. find with -delete is rated high risk rather than disabled, and shred, wipefs, truncate and pkexec are not handled. Because the agent autonomously reads content it does not control, including fetched web pages, repository files and issue text, an indirect prompt injection in that content can cause an unattended run to destroy the invoking user's data. | 12 Oct |
| CVE-2026-62847 | 8.2 high | — | metalama | Metalama is a Roslyn-based open-source patterns and architecture toolkit for C#. Prior to 2025.1.18, 2026.0.25 and 2026.1.19, Metalama.Framework and Metalama.Tool load and execute compile-time assemblies, auxiliary executable tools and embedded helper assemblies from predictable temporary directories. On `Linux` and `macOS`, the default `Path.GetTempPath()` location is world-writable, and existing artifacts are trusted based on sentinel or `.completed` marker files rather than integrity verification. A local unprivileged user on a shared machine or CI/build agent can pre-create the deterministic directory structure and plant a malicious DLL with the matching sentinel. When a higher-privileged developer or build account subsequently builds, Metalama skips regeneration and executes the planted artifact in the compiler, IDE or MSBuild process as the building user. This permits cross-user privilege escalation, reading source and secrets, supply-chain tampering with build output, or disrupting builds. Default Windows per-user temporary directories are protected by ACLs and are not affected. Deployer-selected insecure cache locations are outside the vulnerability's scope. This issue is fixed in versions 2025.1.18, 2026.0.25 and 2026.1.19. | 2 h ago |
| CVE-2026-62331 | 8.2 high | — | mtzanidakis, praktor | praktor is a multi-agent Claude Code orchestrator with Telegram input and output, Docker isolation, and a Mission Control UI. Prior to 0.9.1, both Telegram upload sites in `internal/telegram/bot.go` concatenate the attacker-controlled document filename into a destination path without sanitization, and `path.Join` collapses parent-directory traversal. A remote Telegram user who can message the bot can therefore write outside `uploads/` within the persistent agent workspace mounted at `/workspace/agent`. Because `allow_from: []` permits all senders by default, no prior authentication or privilege is required in that configuration. The attacker can plant or overwrite `CLAUDE.md`, instruction, skill, hook, or configuration files that the Claude agent reads and operates on across later runs, turning a document upload into stored instruction injection. The demonstrated arbitrary-write scope is the persistent agent workspace volume, not an unrestricted host-filesystem write. This issue is fixed in version 0.9.1. | 4 h ago |
| CVE-2026-14467 | 8.2 high | — | forocoches, forocoches comments editor | An information disclosure vulnerability in the Forocoches comments editor that allows an attacker to obtain unauthorised user information by inserting an image containing a malicious external URL into a comment. When a victim views the discussion thread, their browser automatically sends an HTTP/HTTPS request to the server controlled by the attacker, who can then collect information such as the source IP address, the User-Agent header, the date and time of access, and other metadata derived from the request, thereby compromising the user’s privacy. | 7 h ago |
| CVE-2026-42260 | 8.2 high | 0.00% | no CPE yet | Open-WebSearch is a multi-engine MCP server, CLI, and local daemon for agent web search and content retrieval. Prior to 2.1.7, isPublicHttpUrl / assertPublicHttpUrl in src/utils/urlSafety.ts do not recognize bracketed IPv6 literals and do not resolve DNS, which combine to allow non-blind SSRF with the response body returned to the caller. This vulnerability is fixed in 2.1.7. | Mon 26 Oct |
| CVE-2026-43993 | 8.2 high | 0.00% | no CPE yet | JunoClaw is an agentic AI platform built on Juno Network. Prior to 0.x.y-security-1, the WAVS bridge's computeDataVerify called fetch() on agent-supplied URLs without validating scheme, port, or resolved IP, resulting in an SSRF vulnerability. This vulnerability is fixed in 0.x.y-security-1. | Mon 26 Oct |
| CVE-2026-44712 | 8.2 high | 0.00% | no CPE yet | pam_usb provides hardware authentication for Linux using ordinary removable media. Prior to 0.8.7, a crafted UUID such as $(id>/tmp/rce) in the config causes root RCE when pamusb-conf --reset-pads is run. A USB device with a crafted filesystem UUID (some controllers allow this) can inject the payload at --add-device time. Also, userName from the XML config is passed to os.system() in pamusb-agent, which invokes a shell. This vulnerability is fixed in 0.8.7. | Mon 26 Oct |
| CVE-2026-2379 | 8.2 high | 0.00% | no CPE yet | On affected platforms with hardware IPSec support running Arista EOS with certain IPsec features enabled, EOS may exhibit unexpected behavior in specific cases. Physical interface flaps and certain agent restarts can cause IPsec tunnel re-establishment with existing Security Associations, resulting in sequence number mismatches between tunnel endpoints potentially causing unstable communication. | Mon 26 Oct |
| CVE-2026-46866 | 8.2 high | 0.00% | oracle, enterprise_manager_base_platform | Vulnerability in the Oracle Enterprise Manager Base Platform product of Oracle Enterprise Manager (component: Agent Next Gen). Supported versions that are affected are 13.5 and 24.1. Easily exploitable vulnerability allows unauthenticated attacker with network access via HTTPS to compromise Oracle Enterprise Manager Base Platform. Successful attacks of this vulnerability can result in unauthorized ability to cause a hang or frequently repeatable crash (complete DOS) of Oracle Enterprise Manager Base Platform as well as unauthorized update, insert or delete access to some of Oracle Enterprise Manager Base Platform accessible data. CVSS 3.1 Base Score 8.2 (Integrity and Availability impacts). CVSS Vector: (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:H). | Sun 25 Oct |
| CVE-2026-53657 | 8.2 high | 0.00% | no CPE yet | Lima launches Linux virtual machines, typically on macOS, for running containerd. Prior to 2.1.3, on an instance of Lima running with the qemu driver, an arbitrary user in the VM could access /run/lima-guestagent.sock when the guest agent is enabled, which could result in running arbitrary commands with root privileges in the VM because the guest agent socket provides tunneling for arbitrary addresses, including Unix socket addresses for privileged daemons like D-Bus. This issue is fixed in version 2.1.3. | Sun 25 Oct |
| CVE-2026-13410 | 8.2 high | 0.00% | no CPE yet | Dancer::Plugin::Auth::Google versions before 0.08 for Perl have TLS verification disabled. The default user agent is initialised with SSL_verify_mode explicitly disabled. An attacker with network man-in-the-middle (MITM) capability between the Dancer application and googleapis.com can intercept the OAuth2 token exchange and userinfo fetch, return a forged access_token and user profile, and be logged in to the Dancer application as any Google user. | Sat 24 Oct |
| CVE-2026-46993 | 8.2 high | 0.00% | oracle, enterprise_manager_base_platform | Vulnerability in the Oracle Enterprise Manager Base Platform product of Oracle Enterprise Manager (component: Agent Next Gen). Supported versions that are affected are 13.5 and 24.1. Difficult to exploit vulnerability allows low privileged attacker with network access via HTTPS to compromise Oracle Enterprise Manager Base Platform. While the vulnerability is in Oracle Enterprise Manager Base Platform, attacks may significantly impact additional products (scope change). Successful attacks of this vulnerability can result in unauthorized creation, deletion or modification access to critical data or all Oracle Enterprise Manager Base Platform accessible data as well as unauthorized access to critical data or complete access to all Oracle Enterprise Manager Base Platform accessible data. CVSS 3.1 Base Score 8.2 (Confidentiality and Integrity impacts). CVSS Vector: (CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:C/C:H/I:H/A:N). | Sat 24 Oct |
| CVE-2026-56836 | 8.2 high | 0.00% | no CPE yet | PraisonAI is a multi-agent teams system. From version 4.5.112 to before version 4.6.59, PraisonAI's installed console entrypoint is Typer-first. In current releases, the recipe command is registered in the Typer app and praisonai recipe serve dispatches to the deprecated Typer command in src/praisonai/praisonai/cli/commands/recipe.py. That Typer command can start the Recipe HTTP server on a non-localhost interface with no authentication. It prints a deprecation warning, and then launches the server. Because config.auth is absent, create_app() does not attach the API-key or JWT middleware. Unauthenticated requests can then reach the recipe API and, when enabled, /admin/reload. This issue has been patched in version 4.6.59. | 23 Oct |
| CVE-2025-13609 | 8.2 high | 0.00% | no CPE yet | A vulnerability has been identified in keylime where an attacker can exploit this flaw by registering a new agent using a different Trusted Platform Module (TPM) device but claiming an existing agent's unique identifier (UUID). This action overwrites the legitimate agent's identity, enabling the attacker to impersonate the compromised agent and potentially bypass security controls. | 18 Oct |
| CVE-2026-13224 | 8.2 high | 0.00% | watchguard, fireware | A path traversal vulnerability in the Fireware OS WebUI management agent allows an authenticated administrator to read or list arbitrary files on the local filesystem by sending a specially crafted management request. | 8 Oct |
| CVE-2026-40113 | 8.1 high | 0.00% | praison, praisonai | PraisonAI is a multi-agent teams system. Prior to 4.5.128, deploy.py constructs a single comma-delimited string for the gcloud run deploy --set-env-vars argument by directly interpolating openai_model, openai_key, and openai_base without validating that these values do not contain commas. gcloud uses a comma as the key-value pair separator for --set-env-vars. A comma in any of the three values causes gcloud to parse the trailing text as additional KEY=VALUE definitions, injecting arbitrary environment variables into the deployed Cloud Run service. This vulnerability is fixed in 4.5.128. | Mon 26 Oct |
| CVE-2026-39307 | 8.1 high | 0.00% | praison, praisonai | PraisonAI is a multi-agent teams system. Prior to 1.5.113, The PraisonAI templates installation feature is vulnerable to a "Zip Slip" Arbitrary File Write attack. When downloading and extracting template archives from external sources (e.g., GitHub), the application uses Python's zipfile.extractall() without verifying if the files within the archive resolve outside of the intended extraction directory. This vulnerability is fixed in 1.5.113. | Mon 26 Oct |
| CVE-2026-41496 | 8.1 high | 0.00% | praison, praisonai, praisonaiagents | PraisonAI is a multi-agent teams system. Prior to praisonai version 4.6.9 and praisonaiagents version 1.6.9, the fix for CVE-2026-40315 added input validation to SQLiteConversationStore only. Nine sibling backends — MySQL, PostgreSQL, async SQLite/MySQL/PostgreSQL, Turso, SingleStore, Supabase, SurrealDB — pass table_prefix straight into f-string SQL. Same root cause, same code pattern, same exploitation. 52 unvalidated injection points across the codebase. postgres.py additionally accepts an unvalidated schema parameter used directly in DDL. This issue has been patched in praisonai version 4.6.9 and praisonaiagents version 1.6.9. | Mon 26 Oct |
| CVE-2026-47406 | 8.1 high | 0.00% | no CPE yet | PraisonAI Platform is the platform layer for the PraisonAI multi-agent teams system. Versions prior to 0.1.4 have an Insecure Direct Object Reference. The dependency endpoints (`POST/GET /workspaces/{workspace_id}/issues/{issue_id}/dependencies` and `DELETE .../dependencies/{dep_id}`) gate access on `require_workspace_member(workspace_id)` only, then dispatch to `DependencyService` calls that take URL/body-supplied issue and dependency IDs without verifying any of them belong to the membership-checked workspace. Most damaging: `create_dependency` accepts `body.depends_on_issue_id` from the request body — that ID is checked against nothing — letting an attacker create a "blocks" or "related" link between any two issues anywhere in the database. PraisonAI Platform version 0.1.4 patches the issue. | Sat 24 Oct |
| CVE-2026-47409 | 8.1 high | 0.00% | no CPE yet | PraisonAI Platform is the platform layer for the PraisonAI multi-agent teams system. Versions prior to 0.1.4 have an authorization bypass enabling owner lockout. The `DELETE /workspaces/{workspace_id}/members/{user_id}` endpoint is gated only by `require_workspace_member(workspace_id)` (default `min_role="member"`). Any member can remove any other member, including the workspace owner, using a single DELETE. There is no caller-role check, no target-role check, no "cannot remove last owner" guard. PraisonAI Platform version 0.1.4 patches the issue. | Sat 24 Oct |
Download these 806 as CSV · subscribe to this list.
EPSS scores: 2026-10-29. NVD cursor: 2026-10-29 00:22. The News page shows only notable CVEs (KEV, your stack, new criticals, EPSS over 50%, or active exploitation per SSVC); this table has all of them.