CVE-2025-58762
CriticalCVSS 9.1Exploitation Probability (EPSS)
Elevated risk57th percentile - higher than 57% of all known CVEs
Summary
In Tautulli v2.15.3 and earlier, an attacker with administrative access can use the `pms_image_proxy` endpoint to write arbitrary Python scripts into the application filesystem. By changing the PMS URL to a server they control and abusing the `img` and `img_format` parameters (via path traversal), the attacker writes a malicious file and then executes it through the built-in `Script` notification agent, leading to remote code execution.
Risk Assessment
This vulnerability enables remote code execution on the application server, potentially leading to full system compromise, data theft, or further lateral movement within the organization's network. However, it requires administrative privileges in Tautulli.
Recommendation
Upgrade Tautulli to version 2.16.0, which contains the patch. Additionally, restrict administrative access to the application and monitor for unusual PMS configuration changes.
Other vulnerabilities in Tautulli
See all- CVE-2026-43986Critical
Tautulli versions prior to 2.17.1 expose a public `/image/<hash>` route that resolves attacker-controlled entries from `image_hash_lookup` and replays them through server-side image fetch logic. A low-privilege guest user can seed a malicious external image URL into this lookup table and trigger server-side fetches through an unauthenticated endpoint, turning an authenticated SSRF primitive into a persistent unauthenticated SSRF gadget.
- CVE-2026-54915Medium
In Tautulli prior to version 2.17.2, the unauthenticated /auth/redirect endpoint strips forward slashes from the redirect_uri parameter but leaves tab, line-feed, and carriage-return characters intact. With the default HTTP_ROOT configuration this allows redirecting a user to an attacker-controlled external origin.
- CVE-2026-52835High
Tautulli prior to 2.17.2 in the import_config handler and the database_file branch of import_database joins the attacker-controlled filename directly to CACHE_DIR without basename reduction or a containment check. An administrator or caller with the instance API key can submit a multipart filename containing parent-directory segments, causing the upload to be created or overwritten outside CACHE_DIR.
- CVE-2026-49995Medium
In Tautulli prior to version 2.17.2, the newsletter cron field is inserted into a JavaScript string without safe JSON encoding. An administrator or a caller with the Tautulli API key can store a crafted cron value that executes script in the Tautulli web context when an administrator opens the newsletter configuration modal.
- CVE-2026-45381Medium
In Tautulli prior to version 2.17.2, the /search endpoint inserts its user-controlled query parameter into a JavaScript string using manual escaping that does not handle backslashes. A backslash-quote sequence can terminate the string, allowing an unauthenticated attacker to execute script in the Tautulli web context when an authenticated user follows a crafted link.
- CVE-2026-43985High
Tautulli versions prior to 2.17.1 do not enforce POST method on the `configUpdate` endpoint and lack anti-CSRF tokens. An attacker can trick a logged-in administrator into visiting a malicious page that submits a cross-site request, overwriting the admin username and password, leading to administrative interface takeover.
- CVE-2026-43984High
Tautulli is a monitoring tool for Plex Media Server. Prior to 2.17.1, the `log_js_errors` endpoint is accessible to any authenticated user, including guests. It writes attacker-controlled strings to the application log. The admin-only `logFile` view reads the log and embeds it into HTML without escaping, causing stored XSS. A guest can inject HTML/JavaScript that executes in an administrator's browser.
- CVE-2026-41065High
Tautulli prior to 2.17.1 is vulnerable to remote code execution via the newsletter custom template directory feature. On a fresh install before setup wizard completion, all management endpoints are unauthenticated. An attacker can create a newsletter agent, point the template directory to a malicious SMB share with a Mako template, and trigger execution via the newsletter render endpoint.
- CVE-2026-40605Medium
Tautulli before version 2.17.1 has a path traversal vulnerability in the cache deletion endpoint. An authenticated attacker can delete directories outside the configured cache path, leading to data loss and service disruption. Fixed in version 2.17.1.
- CVE-2025-58763High
Tautulli v2.15.3 and prior contains a command injection vulnerability that allows attackers with administrative privileges to achieve remote code execution on the application server. The vulnerability exists when the application is cloned from GitHub and installed manually, as the `runGit` function in `versioncheck.py` uses `shell=True` in `subprocess.Popen` and user-supplied remote and branch names are not sanitized. The fix is in version 2.16.0.
Original NVD description (English source)
Tautulli is a Python based monitoring and tracking tool for Plex Media Server. In Tautulli v2.15.3 and earlier, an attacker with administrative access can use the `pms_image_proxy` endpoint to write arbitrary python scripts into the application filesystem. This leads to remote code execution when combined with the `Script` notification agent. If an attacker with administrative access changes the URL of the PMS to a server they control, they can then abuse the `pms_image_proxy` to obtain a file write into the application filesystem. This can be done by making a `pms_image_proxy` request with a URL in the `img` parameter and the desired file name in the `img_format` parameter. Tautulli then uses a hash of the desired metadata together with the `img_format` in order to construct a file path. Since the attacker controls `img_format` which occupies the end of the file path, and `img_format` is not sanitised, the attacker can then use path traversal characters to specify filename of their choosing. If the specified file does not exist, Tautaulli will then attempt to fetch the image from the configured PMS. Since the attacker controls the PMS, they can return arbitrary content in response to this request, which will then be written into the specified file. An attacker can write an arbitrary python script into a location on the application file system. The attacker can then make use of the built-in `Script` notification agent to run the local script, obtaining remote code execution on the application server. Users should upgrade to version 2.16.0 to receive a patch.
Vulnerability data from NVD (NIST) · CISA KEV · EPSS

