Trigger.dev CVE-2026-114065

Requires a valid authenticated account on a multi-tenant Trigger.dev deployment running a version before 4.7.3 (PR:L). The attacker must send a crafted request to the specific endpoint POST /api/v1/query using the 'FROM {database.table}' placeholder table syntax - including within subqueries - which bypasses the table allowlist and tenant filtering logic. Exploitation is limited to confidentiality: it exposes run metadata and payloads but does not permit modification or code execution. The practical risk is highest for shared/SaaS or multi-tenant self-hosted instances where multiple distinct tenants coexist; a single-tenant deployment has no other tenant data to disclose.

This is a genuine but moderate-priority issue rather than an emergency. The CVSS 4.0 vector (AV:N/AC:L/AT:N/PR:L/UI:N/VC:H) describes a remotely reachable, low-complexity attack that requires only low-level authentication (PR:L) and yields high confidentiality impact with no integrity or availability impact - consistent with a read-only cross-tenant data disclosure. The 7.1 score aligns with this profile. The strongest aggravating signals are the low attack complexity and lack of required user interaction: any authenticated tenant can reach the vulnerable endpoint. However, the key mitigating signals are the authentication requirement and the confidentiality-only impact (no code execution, no data modification). No EPSS score, no CISA KEV listing, and no public POC were provided in the input, so there is no public exploit identified at time of analysis - this is NOT confirmed active exploitation. The absence of EPSS/KEV data is a gap, not evidence of safety; it simply means exploitation likelihood cannot be quantified from the provided sources. Note also a classification nuance: this is tagged and labeled CWE-89 (SQL Injection), but the described mechanism is a tenant-isolation/authorization bypass via placeholder table syntax rather than classic injection of attacker-controlled SQL for execution - the practical impact is unauthorized data read, which the confidentiality-only vector correctly reflects.

An attacker who holds any valid low-privilege account on a multi-tenant Trigger.dev instance sends a POST request to /api/v1/query containing a 'FROM {database.table}' placeholder expression (including inside a subquery) to bypass the table allowlist and tenant filters. The response returns run metadata and task payloads belonging to other tenants, exposing potentially sensitive job data. No public exploit code was identified at time of analysis, and low attack complexity means the technique is straightforward to reproduce once the syntax is known.