Summary
Trigger.dev contains a path traversal and authorization weakness in its packet and object-store handling. A caller with a valid environment API key may access another tenant's object-store objects and potentially read or overwrite task payloads when crafted path segments are normalized before signing.
Affected context
The issue occurs in Trigger.dev deployments using the affected packet API and object-store integration. Risk is present when user-controlled packet keys are incorporated into URL paths, dot segments are accepted, and resource lookup does not verify ownership for the specific object being accessed.
Exploit preconditions
- Attack surface: The packet API and object-store presigning functionality
- Required access: A valid environment API key
- Trigger point: Processing a packet path containing traversal-style dot segments
- Tenant condition: The deployment must contain object-store resources belonging to another tenant
- Why exploitation is possible: URL normalization changes the path before signing, while resource lookup uses an unconditional success result instead of validating per-resource ownership
Why this happens
The vulnerability combines improper input validation (CWE-20), path traversal (CWE-22), and missing authorization (CWE-862). User-controlled packet keys are assigned to URL path components without rejecting dot segments, allowing WHATWG URL normalization to collapse path traversal elements before a presigned request is generated. At the same time, the resource lookup path does not enforce ownership for the requested object, so a valid environment credential can be used across tenant boundaries.
Potential impact
- An attacker may obtain presigned URLs for another tenant's objects.
- An attacker could read confidential task payloads and stored data.
- An attacker could overwrite task payloads or object-store content.
- Cross-tenant data integrity and confidentiality may be compromised.
- Presigned access could enable further abuse of downstream workflows.
General mitigation guidance
Mitigation generally focuses on treating object keys as untrusted identifiers and enforcing authorization at the resource boundary.
Common defensive patterns include:
- Rejecting or canonicalizing dot segments before constructing or signing object-store paths.
- Allowing only validated, schema-conforming object-key formats.
- Resolving paths within an intended tenant- and resource-specific namespace.
- Performing explicit per-resource ownership and capability checks before issuing presigned URLs.
- Avoiding unconditional resource-lookup callbacks that report authorization success.
- Applying least-privilege permissions to object-store credentials and presigned operations.
- Using secure-by-default APIs that separate path construction from user-controlled input.
How teams detect issues like this
Organizations can detect similar issues through:
- Static analysis for user input assigned directly to URL paths, filesystem paths, or object keys.
- Code review checks for missing rejection of dot segments and inconsistent path canonicalization.
- Review of presigned URL generation to confirm authorization occurs before signing.
- Tests that exercise cross-tenant identifiers and verify read and write isolation.
- Policy enforcement requiring explicit ownership checks for resource access.
- Monitoring for unusual presigned URL requests, repeated traversal-like paths, and cross-tenant access patterns.
Automated security review tools may help identify path construction and authorization gaps, but should be supplemented with tenant-isolation testing.

