Trigger.dev CVE-2026-114082

Requires the attacker to already possess a public token that has session write access to the target Trigger.dev environment (PR:L); with that token, exploitation is performed over the network by using the session input initialization path to retrieve environment-wide S2 stream credentials, which are then used directly against the storage API. Limiting factors: it is NOT unauthenticated - a valid session-write-capable public token is a hard prerequisite - and affected deployments must be running a version before 4.6.2; no user interaction and no non-default feature toggle is indicated in the description.

This is a genuine but access-gated integrity/availability risk rather than a mass-exploitation emergency. The supplied CVSS 4.0 vector (AV:N/AC:L/AT:N/PR:L/UI:N) and the CWE-862 (Missing Authorization) classification agree with the description: exploitation is network-reachable with low complexity but requires PR:L - the attacker must already hold a valid public token with session write access, so this is privileged-insider or compromised-client abuse, not anonymous internet exploitation. Impact is skewed to integrity and availability (VI:H/VA:H) with no confidentiality loss (VC:N), consistent with forging/trimming/deleting other sessions' and runs' streams. No EPSS score, CISA KEV listing, or public POC was provided in the input, so there is no public exploit identified at time of analysis and no evidence of active exploitation - the 7.2 'High' rating is driven by the broad write/delete capability once the precondition is met, not by observed in-the-wild activity. One point to verify with the vendor: the stolen credential is described as 'environment-wide,' which implies cross-session/cross-run impact, yet the 4.0 vector sets subsequent-system metrics to N (SC/SI/SA:N); defenders should confirm whether 'environment-wide' constitutes a security-authority/scope change.

An attacker who holds a low-privilege public token with session write access (e.g., a compromised or over-scoped frontend client) initializes a session's input and is handed environment-wide S2 stream credentials. Using that credential directly against the storage API, they append forged output to another user's run stream and trim or delete legitimate stream records. No public exploit code was identified in the provided data, and exploitation requires only network access plus the existing low-privilege token (AC:L, no user interaction).