You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A known sensitive canary file was accessed in Google Workspace, indicating attempts to access sensitive information or files without proper permissions.
This would require the prerequisite step of strategically naming and placing the canary files in locations that are deemed the most vulnerable (or interesting). Ideally, this would be a rule vs a hunt, however, because of the unique nature of all environments and canary naming patterns, it is not easily done via a prebuilt rule.
Where this get's really interesting - expanding beyond this rule
To take this detection even further, an accompanying OSQuery and/or YARA signature can be created to detect when the file downloaded or opened
Steps:
create this hunt (or better, custom rule), defining sensitive file name, path, or other criteria
(optionally) create an OSQuery recurring search to detect it on name or hash
SELECT*FROM file_events
WHERE action ='opened'AND (
target_path IN (
'C:\\path\\to\\your\\sensitive.doc', -- Windows path'/path/to/your/sensitive.doc', -- Linux path'/path/to/your/sensitive.doc'-- macOS path
)
OR
sha256 ='<computed_hash_value>'
);
(optionally) create an OSQuery recurring search to run a YARA scan and write a YARA signature to look for identifying patterns
Target Huntset
google_workspace
Target hunt Type
ES|QL
Query
from logs-google_workspace*
| wherefile.name=="*"orfile.nameis not nullandevent.actionin ("view", "download") and
(
//expand to canary patterns
file.namein ("known", "canary", "pattterns")
)
| stats files=count(*) by file.name, user.email, event.action
| sort files asc
Related issues or PRs
No response
References
No response
Redacted Example Data
No response
The text was updated successfully, but these errors were encountered:
Description
A known sensitive canary file was accessed in Google Workspace, indicating attempts to access sensitive information or files without proper permissions.
This would require the prerequisite step of strategically naming and placing the canary files in locations that are deemed the most vulnerable (or interesting). Ideally, this would be a rule vs a hunt, however, because of the unique nature of all environments and canary naming patterns, it is not easily done via a prebuilt rule.
Where this get's really interesting - expanding beyond this rule
To take this detection even further, an accompanying OSQuery and/or YARA signature can be created to detect when the file downloaded or opened
Steps:
Target Huntset
google_workspace
Target hunt Type
ES|QL
Query
Related issues or PRs
No response
References
No response
Redacted Example Data
No response
The text was updated successfully, but these errors were encountered: