| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| IBM DataStage on Cloud Pak for Data 5.4.0.0 could allow a remote authenticated attacker to execute arbitrary commands due to improper neutralization of special elements used in an OS command. |
| IBM DataStage on Cloud Pak for Data 5.4.0.0 could allow a remote authenticated attacker to execute arbitrary commands due to path traversal. |
| IBM DataStage on Cloud Pak for Data 5.4.0.0 could allow a remote authenticated attacker to execute arbitrary code due to improper neutralization of special elements used in an OS command. |
| MoguBlog through 6.2 contains an authorization bypass vulnerability in the comment deletion endpoint that performs ownership checks against request-body fields instead of the authenticated principal. Attackers can delete arbitrary comments and their replies by supplying comment UIDs and author UIDs obtained from unauthenticated listing endpoints. |
| IBM ContextForge MCP Gateway 1.0.0 through 1.0.8 was vulnerable to path traversal in its Admin API log-download endpoint (`GET /v1/admin/logs/file`). The path confinement check uses `str.startswith()` rather than proper boundary validation, allowing an authenticated admin to read `.log`, `.jsonl`, and `.json` files outside the configured `LOG_FOLDER` by supplying a filename that resolves into a sibling directory whose absolute path shares the log directory's string prefix. |
| SQL Chat contains four unauthenticated API endpoints that accept client-supplied database connection parameters and execute arbitrary SQL queries against attacker-specified hosts. Attackers can connect to internal databases, execute SQL commands, enumerate schemas, and pivot into the server's network without authentication. |
| claude-skill-antivirus fails to analyze executable files when scanning local skill directories, reading only SKILL.md while ignoring Python source, bytecode, and other artifacts in the scripts directory. Attackers can distribute skills with malicious code in non-manifest files that receive a SAFE verdict with 100/100 trust score despite containing unanalyzed executable payloads. |
| argocd-mcp 0.8.0 binds its HTTP transport to every network interface and accepts MCP sessions without requiring caller credentials when ARGOCD_API_TOKEN is configured. Attackers who can reach the listener can invoke the full tool surface using the operator's stored token to create applications, request syncs, and modify Argo CD resources. |
| gpt-crawler through 1.5.1 fails to validate the outputFileName parameter in the POST /crawl endpoint, allowing unauthenticated attackers to write arbitrary files to any filesystem path. Attackers can supply absolute paths or parent-directory segments to overwrite existing files with content sourced from attacker-controlled URLs. |
| Dolibarr 9.0.0 through 23.0.4 saves inbound email attachments under the name supplied in the message's MIME headers without reducing it to a safe basename. The global saveAttachment() in htdocs/emailcollector/lib/emailcollector.lib.php builds $filepath = $path . $filename . '.' . $ext and hands it to file_put_contents(), and the private saveAttachment() in htdocs/emailcollector/class/emailcollector.class.php writes to $destdir.'/'.$filename; the name reaches both from the attachment's own getName() or getFilename() value by way of the record-join, create-ticket and create-project operations. A traversal sequence in the filename therefore survives intact, so any sender who can email a mailbox that an EmailCollector monitors, which is the module's ordinary use for a support or ticket inbox, can place attacker-controlled content outside the per-object attachment directory without holding a Dolibarr account. Under the hardened layout Dolibarr's SECURITY.md requires, with htdocs read-only, the write is confined to the documents tree and corrupts or forges other objects' documents; where htdocs is writable the same primitive reaches a web-executable path. Version 24.0.0 applies dol_sanitizePathName() and dol_sanitizeFileName() before the write. |
| The updateWorkspace handler in mods/identity/src/workspaces/createUpdateWorkspace.ts in Fonoster through 0.22.7 invokes the gRPC callback with PERMISSION_DENIED when createIsWorkspaceMember reports that the caller is not a member of the target workspace, but it does not return. Execution continues into prisma.workspace.update, which is scoped by the workspace reference alone, so the rename commits before the second callback is issued and the caller receives a permission error for a write that already succeeded. The gRPC interceptor in mods/common/src/identity/createAuthInterceptor.ts binds the workspace accessKeyId to the caller's token only for paths in workspaceResourceAccess or workspaceResourceOwnerOrAdminAccess, and Identity/UpdateWorkspace is listed in fullIdentityAccess, which the base USER role holds. Any authenticated user can therefore rename an arbitrary workspace in the deployment. |
| Pydio Cells 5.0.0 through 5.0.2 returns share-link details to any authenticated user. The REST handler for GET /a/share/link/{Uuid} in idm/share/rest/handler.go reads the workspace UUID from the path, calls LinkById, and writes the result with no authorization step, whereas the sibling handler for GET /a/share/cell/{Uuid} loads the workspace and requires MatchPolicies with ResourcePolicyAction_READ, returning a not-found error so that existence is not disclosed. Nothing compensates further down: GetLinkWorkspace reaches GetOrCreateWorkspace, which issues SearchWorkspace with a query carrying no ResourcePolicyQuery, and PrepareResourcePolicyQuery returns the query unmodified when that field is nil, so the workspace service applies no policy filter. The workspace UUID is not secret, because the unauthenticated public page served for a share link embeds it as START_REPOSITORY. Any account holding a standard user role can therefore submit the UUID and receive the link hash and URL, the owner's user identifier, the hidden share user login, the permission set, the download limit and count, the target users, the expiry, and the password-required flag, while a direct read of the shared node from the same account is refused. |
| Evidently UI fails to properly validate the filename parameter in the dataset materialization endpoint, allowing unauthenticated attackers to read arbitrary files outside the workspace directory. Attackers can supply traversal sequences or absolute paths in the filename field to access system files, which are then materialized into datasets and retrieved through the download endpoint. |
| Cockpit CMS 2.14.0 and prior contains a command injection vulnerability in the FFmpeg integration that allows authenticated users with only the assets/upload permission to execute arbitrary commands by uploading a video file with a shell metacharacter-laden filename. The unsanitized filename is interpolated into a shell command executed via Process::fromShellCommandline() before the slugify() sanitizer runs, enabling injected shell metacharacters such as backticks, $(), and semicolons to escape the FFmpeg command context and execute as the web-server user. |
| ImpressCMS contains an authenticated remote code execution vulnerability in the custom tag module that allows authenticated administrators to execute arbitrary PHP code by storing a malicious payload in a custom tag with PHP type enabled. The application decodes HTML-encoded content via undoHtmlSpecialChars() before passing it to eval() in the renderWithPhp() method, bypassing HTML Purifier sanitization, and the payload is triggered on every frontend page load through the preload event system. |
| MindsDB Minds Platform version 26.1.0 and earlier contains an unauthenticated remote code execution vulnerability that allows unauthenticated attackers to execute arbitrary OS commands by submitting crafted prompts to the unprotected POST /api/v1/responses/ endpoint, which reaches the Anton agent's scratchpad tool that calls exec() on attacker-influenced Python source without sandboxing. Attackers can first configure their own LLM API key through the unauthenticated PUT /api/v1/settings/ endpoint, then POST a prompt directing the agent to invoke the scratchpad tool with arbitrary Python code, achieving full OS command execution as the user running the desktop application and enabling access to SSH keys, stored credentials, and environment secrets. |
| dua-cli fails to filter terminal escape sequences when printing marked file paths after exiting the TUI interface. Attackers can craft file names containing OSC/CSI escape sequences that are interpreted by the terminal emulator when printed, enabling title spoofing, clipboard manipulation, or other escape-sequence attacks. |
| The github-issue-bot templates in appwrite/templates verify the GitHub webhook signature with an inverted condition. verifyWebhook in node/github-issue-bot/src/github.js and in node-typescript/github-issue-bot/src/github.ts returns "typeof signature !== 'string' || (await verify(...))", so when the X-Hub-Signature-256 header is absent the first operand is true, the logical OR short-circuits, and the function reports success without performing any HMAC verification. main.js rejects a request only when verifyWebhook returns false, so an unauthenticated request carrying no signature passes the check. Processing then continues to postComment, which takes the repository and issue objects directly from the request body, letting the caller direct the deployed function to post a comment on a repository and issue of their choosing using the configured GITHUB_TOKEN, with the issue author login from the body interpolated into the comment text. |
| broot renders each file and directory name in its interactive tree view exactly as read from the filesystem. Names are converted with a plain to_string_lossy() call in src/tree_build/builder.rs and in TreeLine::unprune in src/tree/tree_line.rs, and no control-character filtering exists anywhere in the code, even though the doc comment on the TreeLine name field states that some characters may have been stripped. Any local user who can create a file can therefore place an escape sequence in its name and have it written unmodified to the terminal of anyone who browses that directory, between broot's own styling codes. A reported proof of concept used an OSC 52 clipboard-write sequence and captured the raw bytes broot wrote to its pty, confirming the sequence reaches the terminal unstripped. What an injected OSC or CSI sequence can then do depends on the terminal emulator in use. Browsing a directory is broot's primary function and carries no expectation that the content is trusted. |
| The customer update route in EverShop is declared with "access": "public" in packages/evershop/src/modules/customer/api/updateCustomer/route.json, which causes the admin authentication middleware to call next() without checking the caller, and no customer-session middleware guards the route; the only middleware in the chain parses the JSON body. The handler in updateCustomer.js then loads the customer by the uuid taken from the URL path and writes the supplied fields back to that record, hashing a password if one is provided, without verifying that the caller owns the record. An unauthenticated request carrying a known customer uuid can therefore overwrite that customer's email address and password and read back the updated record from the 200 response, taking over the account and locking out its owner. Customer uuids are exposed through order confirmation email links and administrative URLs. Version 2.2.1 changes the route to "access": "private". |
