Vulnerabilities (CVE)

libgphoto2 is a camera access and control library. In versions up to and including 2.5.33, two functions in camlibs/ptp2/ptp-pack.c accept a data pointer but no length parameter, performing unbounded reads. Their callers in ptp_unpack_EOS_events() have xsize available but never pass it, leaving both functions unable to validate reads against the actual buffer boundary. Commit 1817ecead20c2aafa7549dac9619fe38f47b2f53 patches the issue.Show less

Hot Chocolate is an open-source GraphQL server. Prior to versions 12.22.7, 13.9.16, 14.3.1, and 15.1.14, Hot Chocolate's recursive descent parser `Utf8GraphQLParser` has no recursion depth limit. A crafted GraphQL document with deeply nested selection sets, object values, list values, or list types can trigger a `StackOverflowException` on payloads as small as 40 KB. Because `StackOverflowException` is uncatchable in .NET (since .NET 2.0), the entire worker process is terminated immediately. All in-flight HTTP requests, background `IHostedService` tasks, and open WebSocket subscriptions on that worker are dropped. The orchestrator (Kubernetes, IIS, etc.) must restart the process. This occurs before any validation rules run — `MaxExecutionDepth`, complexity analyzers, persisted query allow-lists, and custom `IDocumentValidatorRule` implementations cannot intercept the crash because `Utf8GraphQLParser.Parse` is invoked before validation. The `MaxAllowedFields=2048` limit does not help because the crashing payloads contain very few fields. The fix in versions 12.22.7, 13.9.16, 14.3.1, and 15.1.14 adds a `MaxAllowedRecursionDepth` option to `ParserOptions` with a safe default, and enforces it across all recursive parser methods (`ParseSelectionSet`, `ParseValueLiteral`, `ParseObject`, `ParseList`, `ParseTypeReference`, etc.). When the limit is exceeded, a catchable `SyntaxException` is thrown instead of overflowing the stack. There is no application-level workaround. `StackOverflowException` cannot be caught in .NET. The only mitigation is to upgrade to a patched version. Operators can reduce (but not eliminate) risk by limiting HTTP request body size at the reverse proxy or load balancer layer, though the smallest crashing payload (40 KB) is well below most default body size limits and is highly compressible (~few hundred bytes via gzip).Show less

DNN (formerly DotNetNuke) is an open-source web content management platform (CMS) in the Microsoft ecosystem. Prior to version 10.2.2, a user could upload a specially crafted SVG file that could include scripts that can target both authenticated and unauthenticated DNN users. The impact is increased if the scripts are run by a power user. Version 10.2.2 patches the issue.Show less

Giskard is an open-source testing framework for AI models. In versions prior to 1.0.2b1, the ConformityCheck class rendered the rule parameter through Jinja2's default Template() constructor, silently interpreting template expressions at runtime. If check definitions are loaded from an untrusted source, a crafted rule string could achieve arbitrary code execution. Exploitation requires write access to a check definition and subsequent execution of the test suite. This issue has been fixed in giskard-checks version 1.0.2b1.Show less

OWASP BLT is a QA testing and vulnerability disclosure platform that encompasses websites, apps, git repositories, and more. Versions prior to 2.1.1 contain an RCE vulnerability in the .github/workflows/regenerate-migrations.yml workflow. The workflow uses the pull_request_target trigger to run with full GITHUB_TOKEN write permissions, copies attacker-controlled files from untrusted pull requests into the trusted runner workspace via git show, and then executes python manage.py makemigrations, which imports Django model modules including attacker-controlled website/models.py at runtime. Any module-level Python code in the attacker's models.py is executed during import, enabling arbitrary code execution in the privileged CI environment with access to GITHUB_TOKEN and repository secrets. The attack is triggerable by any external contributor who can open a pull request, provided a maintainer applies the regenerate-migrations label, potentially leading to secret exfiltration, repository compromise, and supply chain attacks. A patch for this issue is expected to be released in version 2.1.1.Show less

PraisonAI is a multi-agent teams system. Prior to 4.5.133, there is an SQL identifier injection vulnerability in SQLiteConversationStore where the table_prefix configuration value is directly concatenated into SQL queries via f-strings without any validation or sanitization. Since SQL identifiers cannot be safely parameterized, an attacker who controls the table_prefix value (e.g., through from_yaml or from_dict configuration input) can inject arbitrary SQL fragments that alter query structure. This enables unauthorized data access, such as reading internal SQLite tables like sqlite_master, and manipulation of query results through techniques like UNION-based injection. The vulnerability propagates from configuration input in config.py, through factory.py, to the SQL query construction in sqlite.py. Exploitation requires the ability to influence configuration input, and successful exploitation leads to internal schema disclosure and full query result tampering. This issue has been fixed in version 4.5.133.Show less

zrok is software for sharing web services, files, and network resources. Prior to version 2.0.1, the unaccess handler (controller/unaccess.go) contains a logical error in its ownership guard: when a frontend record has environment_id = NULL (the marker for admin-created global frontends), the condition short-circuits to false and allows the deletion to proceed without any ownership verification. A non-admin user who knows a global frontend token can call DELETE /api/v2/unaccess with any of their own environment IDs and permanently delete the global frontend, taking down all public shares routed through it. Version 2.0.1 patches the issue.Show less

zrok is software for sharing web services, files, and network resources. Prior to version 2.0.1, endpoints.GetSessionCookie parses an attacker-supplied cookie chunk count and calls make([]string, count) with no upper bound before any token validation occurs. The function is reached on every request to an OAuth-protected proxy share, allowing an unauthenticated remote attacker to trigger gigabyte-scale heap allocations per request, leading to process-level OOM termination or repeated goroutine panics. Both publicProxy and dynamicProxy are affected. Version 2.0.1 patches the issue.Show less

zrok is software for sharing web services, files, and network resources. Prior to version 2.0.1, the proxyUi template engine uses Go's text/template (which performs no HTML escaping) instead of html/template. The GitHub OAuth callback handlers in both publicProxy and dynamicProxy embed the attacker-controlled refreshInterval query parameter verbatim into an error message when time.ParseDuration fails, and render that error unescaped into HTML. An attacker can deliver a crafted login URL to a victim; after the victim completes the GitHub OAuth flow, the callback page executes arbitrary JavaScript in the OAuth server's origin. Version 2.0.1 patches the issue.Show less

PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the browser bridge (praisonai browser start) is vulnerable to unauthenticated remote session hijacking due to missing authentication and a bypassable origin check on its /ws WebSocket endpoint. The server binds to 0.0.0.0 by default and only validates the Origin header when one is present, meaning any non-browser client that omits the header is accepted without restriction. An unauthenticated network attacker can connect, send a start_session message, and the server will route it to the first idle browser-extension WebSocket (effectively hijacking that session) and then broadcast all resulting automation actions and outputs back to the attacker. This enables unauthorized remote control of connected browser automation sessions, leakage of sensitive page context and automation results, and misuse of model-backed browser actions in any environment where the bridge is network-reachable. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents.Show less

PraisonAI is a multi-agent teams system. In versions below 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents, the workflow engine is vulnerable to arbitrary command and code execution through untrusted YAML files. When praisonai workflow run <file.yaml> loads a YAML file with type: job, the JobWorkflowExecutor in job_workflow.py processes steps that support run: (shell commands via subprocess.run()), script: (inline Python via exec()), and python: (arbitrary Python script execution)—all without any validation, sandboxing, or user confirmation. The affected code paths include action_run() in workflow.py and _exec_shell(), _exec_inline_python(), and _exec_python_script() in job_workflow.py. An attacker who can supply or influence a workflow YAML file (particularly in CI pipelines, shared repositories, or multi-tenant deployment environments) can achieve full arbitrary command execution on the host system, compromising the machine and any accessible data or credentials. This issue has been fixed in versions 4.5.139 of PraisonAI and 1.5.140 of praisonaiagents.Show less

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.Show less

Note Mark is an open-source note-taking application. In versions 0.19.1 and prior, the asset download endpoint at /api/notes/{noteID}/assets/{assetID} is registered without authentication middleware, and the backend query does not verify ownership or book visibility. An unauthenticated user who knows a valid note ID and asset ID can retrieve the full contents of private note assets without authentication, regardless of whether the associated book is public or private. This issue has been fixed in version 0.19.2.Show less

Net::CIDR::Lite versions before 0.23 for Perl mishandles IPv4 mapped IPv6 addresses, which may allow IP ACL bypass. _pack_ipv6() includes the sentinel byte from _pack_ipv4() when building the packed representation of IPv4 mapped addresses like ::ffff:192.168.1.1. This produces an 18 byte value instead of 17 bytes, misaligning the IPv4 part of the address. The wrong length causes incorrect results in mask operations (bitwise AND truncates to the shorter operand) and in find() / bin_find() which use Perl string comparison (lt/gt). This can cause find() to incorrectly match or miss addresses. Example: my $cidr = Net::CIDR::Lite->new("::ffff:192.168.1.0/120"); $cidr->find("::ffff:192.168.2.0"); # incorrectly returns true This is triggered by valid RFC 4291 IPv4 mapped addresses (::ffff:x.x.x.x). See also CVE-2026-40198, a related issue in the same function affecting malformed IPv6 addresses.Show less

HomeBox is a home inventory and organization system. Versions prior to 0.25.0 contain a vulnerability where the defaultGroup ID remained permanently assigned to a user after being invited to a group, even after their access to that group was revoked. While the web interface correctly enforced the access revocation and prevented the user from viewing or modifying the group's contents, the API did not. Because the original group ID persisted as the user's defaultGroup, and this value was not properly validated when the X-Tenant header was omitted, the user could still perform full CRUD operations on the group's collections through the API, bypassing the intended access controls. This issue has been fixed in version 0.25.0.Show less

goshs is a SimpleHTTPServer written in Go. Prior to 2.0.0-beta.4, goshs enforces the documented per-folder .goshs ACL/basic-auth mechanism for directory listings and file reads, but it does not enforce the same authorization checks for state-changing routes. An unauthenticated attacker can upload files with PUT, upload files with multipart POST /upload, create directories with ?mkdir, and delete files with ?delete inside a .goshs-protected directory. By deleting the .goshs file itself, the attacker can remove the folder's auth policy and then access previously protected content without credentials. This results in a critical authorization bypass affecting confidentiality, integrity, and availability. This vulnerability is fixed in 2.0.0-beta.4.Show less

OpenTelemetry dotnet is a dotnet telemetry framework. From 1.13.1 to before 1.15.2, When exporting telemetry to a back-end/collector over gRPC or HTTP using OpenTelemetry Protocol format (OTLP), if the request results in a unsuccessful request (i.e. HTTP 4xx or 5xx), the response is read into memory with no upper-bound on the number of bytes consumed. This could cause memory exhaustion in the consuming application if the configured back-end/collector endpoint is attacker-controlled (or a network attacker can MitM the connection) and an extremely large body is returned by the response. This vulnerability is fixed in 1.15.2.Show less

jq is a command-line JSON processor. Before commit 0c7d133c3c7e37c00b6d46b658a02244fdd3c784, jq used MurmurHash3 with a hardcoded, publicly visible seed (0x432A9843) for all JSON object hash table operations, which allowed an attacker to precompute key collisions offline. By supplying a crafted JSON object (~100 KB) where all keys hashed to the same bucket, hash table lookups degraded from O(1) to O(n), turning any jq expression into an O(n²) operation and causing significant CPU exhaustion. This affected common jq use cases such as CI/CD pipelines, web services, and data processing scripts, and was far more practical to exploit than existing heap overflow issues since it required only a small payload. This issue has been patched in commit 0c7d133c3c7e37c00b6d46b658a02244fdd3c784.Show less

PraisonAI is a multi-agent teams system. Prior to 4.5.128, PraisonAI’s MCP (Model Context Protocol) integration allows spawning background servers via stdio using user-supplied command strings (e.g., MCP("npx -y @smithery/cli ...")). These commands are executed through Python’s subprocess module. By default, the implementation forwards the entire parent process environment to the spawned subprocess. As a result, any MCP command executed in this manner inherits all environment variables from the host process, including sensitive data such as API keys, authentication tokens, and database credentials. This behavior introduces a security risk when untrusted or third-party commands are used. In common scenarios where MCP tools are invoked via package runners such as npx -y, arbitrary code from external or potentially compromised packages may execute with access to these inherited environment variables. This creates a risk of unintended credential exposure and enables potential supply chain attacks through silent exfiltration of secrets. This vulnerability is fixed in 4.5.128.Show less

PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, he list_files() tool in FileTools validates the directory parameter against workspace boundaries via _validate_path(), but passes the pattern parameter directly to Path.glob() without any validation. Since Python's Path.glob() supports .. path segments, an attacker can use relative path traversal in the glob pattern to enumerate arbitrary files outside the workspace, obtaining file metadata (existence, name, size, timestamps) for any path on the filesystem. This vulnerability is fixed in 1.5.128.Show less

PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, read_skill_file() in skill_tools.py allows reading arbitrary files from the filesystem by accepting an unrestricted skill_path parameter. Unlike file_tools.read_file which enforces workspace boundary confinement, and unlike run_skill_script which requires critical-level approval, read_skill_file has neither protection. An agent influenced by prompt injection can exfiltrate sensitive files without triggering any approval prompt. This vulnerability is fixed in 1.5.128.Show less

PraisonAI is a multi-agent teams system. Prior to 4.5.128, the /media-stream WebSocket endpoint in PraisonAI's call module accepts connections from any client without authentication or Twilio signature validation. Each connection opens an authenticated session to OpenAI's Realtime API using the server's API key. There are no limits on concurrent connections, message rate, or message size, allowing an unauthenticated attacker to exhaust server resources and drain the victim's OpenAI API credits. This vulnerability is fixed in 4.5.128.Show less

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.Show less

PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, he memory hooks executor in praisonaiagents passes a user-controlled command string directly to subprocess.run() with shell=True at src/praisonai-agents/praisonaiagents/memory/hooks.py. No sanitization is performed and shell metacharacters are interpreted by /bin/sh before the intended command executes. Two independent attack surfaces exist. The first is via pre_run_command and post_run_command hook event types registered through the hooks configuration. The second and more severe surface is the .praisonai/hooks.json lifecycle configuration, where hooks registered for events such as BEFORE_TOOL and AFTER_TOOL fire automatically during agent operation. An agent that gains file-write access through prompt injection can overwrite .praisonai/hooks.json and have its payload execute silently at every subsequent lifecycle event without further user interaction. This vulnerability is fixed in 1.5.128.Show less

Flux notification-controller is the event forwarder and notification dispatcher for the GitOps Toolkit controllers. Prior to 1.8.3, the gcr Receiver type in Flux notification-controller does not validate the email claim of Google OIDC tokens used for Pub/Sub push authentication. This allows any valid Google-issued token, to authenticate against the Receiver webhook endpoint, triggering unauthorized Flux reconciliations. Exploitation requires the attacker to know the Receiver's webhook URL. The webhook path is generated as /hook/sha256sum(token+name+namespace), where the token is a random string stored in a Kubernetes Secret. There is no API or endpoint that enumerates webhook URLs. An attacker cannot discover the path without either having access to the cluster and permissions to read the Receiver's .status.webhookPath in the target namespace, or obtaining the URL through other means (e.g. leaked secrets or access to Pub/Sub config). Upon successful authentication, the controller triggers a reconciliation for all resources listed in the Receiver's .spec.resources. However, the practical impact is limited: Flux reconciliation is idempotent, so if the desired state in the configured sources (Git, OCI, Helm) has not changed, the reconciliation results in a no-op with no effect on cluster state. Additionally, Flux controllers deduplicate reconciliation requests, sending many requests in a short period results in only a single reconciliation being processed. This vulnerability is fixed in 1.8.3.Show less