When invoked, the endpoint triggers the execution of email_reader.py as a subprocess. This script automatically attempts to load credentials from config/connector_oauth/.env.connector.oauth and initiates IMAP connections to configured ProtonMail or Gmail accounts. The API server captures the standard output of this process, which includes sender information, subjects, and body snippets of fetched RFC822 messages, and returns this sensitive data to the caller in a JSON response. This provides a mechanism for unauthorized actors to exfiltrate operational intelligence without requiring credentials or prior access to the host environment.

Attack Chain

1. Attacker performs network scanning to identify reachable instances of AetherBrowser listening on port 8100.
2. Attacker crafts a standard HTTP POST request to the /api/ops/check-email endpoint.
3. The api_server.py processes the request, bypassing authentication checks due to the missing Depends() decorator.
4. The server executes email_reader.py as a system subprocess (_run_subprocess).
5. The email_reader.py script parses local environment files to retrieve configured IMAP credentials.
6. The script establishes an IMAP connection to the target mail provider (ProtonMail or Gmail) using the retrieved credentials.
7. The script fetches RFC822 messages, parses them, and writes email metadata to stdout.
8. The API server intercepts the first 2000 characters of stdout and returns the email data to the attacker in the HTTP response.

Impact

Successful exploitation allows unauthenticated remote attackers to exfiltrate email metadata, including sender identities, subjects, and content snippets. This disclosure of sensitive operational communications can facilitate follow-on social engineering or targeted attacks against operators. Additionally, repeated calls to the endpoint can lead to IMAP connection exhaustion or account-level security alerts, effectively disclosing the operational infrastructure's mail configuration and diagnostic status.

Recommendation

1. Patch immediately by implementing mandatory API-key validation using a header-based check (e.g., X-API-KEY) for all /api/ops/* endpoints.
2. Modify the API server configuration to bind only to 127.0.0.1 to prevent remote access unless explicitly required by the architecture.
3. Disable wildcard CORS (allow_origins=["*"]) and restrict access to specific, trusted origins.
4. Implement strict input and output sanitization for subprocess interactions; the application should not return raw stdout from operational scripts to external API callers.
5. Rotate all mail credentials stored in config/connector_oauth/.env.connector.oauth if it is suspected that the API was exposed to untrusted networks.