Summary
AVideo's Live plugin exposes a broadcast callback without verifying that requests come from an authorized source. An unauthenticated attacker may alter the status of arbitrary scheduled broadcasts, causing legitimate live events to be marked as failed or silently canceled.
Affected context
The vulnerability occurs in the AVideo Live plugin's on_publish.php callback, where incoming broadcast-related requests may be processed without sufficient authentication or authorization checks. Deployments are at risk when this callback is reachable by untrusted network clients and schedule identifiers or stream-key-derived values are accepted without verifying ownership of the associated broadcast.
Exploit preconditions
- Attack surface: The Live plugin's RTMP publish callback endpoint
- Required access: Network access to submit requests; no valid account or established session is required
- Trigger point: Processing of a crafted publish callback containing a schedule identifier
- Why exploitation is possible: The endpoint accepts attacker-controlled identifiers and updates scheduled broadcast state without confirming request authenticity or authorization
Why this happens
The underlying flaw is missing authentication and authorization on a state-changing callback. The application treats externally supplied callback data as trusted and permits it to select and modify scheduled broadcasts without verifying that the request originated from a legitimate streaming workflow or that the requester controls the target schedule. This behavior maps to CWE-284: Improper Access Control, including the absence of an effective authentication boundary before a privileged state transition.
Potential impact
- Scheduled broadcasts may be marked as failed without the owner's approval
- Legitimate live events could be canceled or disrupted
- Operators may experience denial of service against broadcast scheduling
- Broadcast availability and operational monitoring may become unreliable
- Repeated unauthorized changes could undermine trust in scheduling controls
General mitigation guidance
Mitigation generally focuses on protecting state-changing callbacks with strong authenticity and authorization controls.
Common defensive patterns include:
- Require authenticated, integrity-protected callbacks from trusted streaming infrastructure
- Verify that the requested schedule belongs to the authenticated publishing context
- Avoid treating predictable or externally supplied identifiers as authorization evidence
- Apply strict validation and replay protections to callback requests
- Limit the callback's permissions to only the broadcast state changes it requires
- Restrict network exposure of administrative and publishing endpoints where practical
These controls reduce the likelihood that an unauthenticated party can manipulate broadcast state.
How teams detect issues like this
Organizations can look for this issue through:
- Static analysis rules identifying state-changing handlers without authentication or authorization gates
- Code review checks for trust decisions based solely on stream keys, identifiers, or request parameters
- Tests that submit callback requests without credentials and verify that no schedule state changes occur
- Monitoring for unexpected broadcast-status transitions, repeated callback failures, or requests from untrusted sources
- Policy enforcement requiring authentication and integrity validation for webhook and streaming callback endpoints
Automated security review tools may help identify exposed handlers and missing access-control checks, but runtime testing and configuration review remain important.

