GHSA-6whj-7qmg-86qj
MEDIUMKhoj has an IDOR in Notion OAuth Flow that Enables Index Poisoning
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
khojReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects PyPI packages — download data is not available via public APIs for these ecosystems.
Description
Summary
An IDOR in the Notion OAuth callback allows an attacker to hijack any user's Notion integration by manipulating the state parameter. The callback endpoint accepts any user UUID without verifying the OAuth flow was initiated by that user, allowing attackers to replace victims' Notion configurations with their own, resulting in data poisoning and unauthorized access to the victim's Khoj search index.
This attack requires knowing the user's UUID which can be leaked through shared conversations where an AI generated image is present.
Details
When users share conversations which contain AI generated images, the file path for the image is constructed using the user's UUID. Knowing this UUID, an attacker is able to intercept the OAuth callback for Notion and replace the state parameter with the other user's UUID and sync notion onto their account.
PoC
The vulnerable line of code exists in src/khoj/routers/notion.py on the callback endpoint.
@notion_router.get("/auth/callback")
async def notion_auth_callback(request: Request, background_tasks: BackgroundTasks):
code = request.query_params.get("code")
state = request.query_params.get("state") # <-- Attacker controlled
if not code or not state:
return Response("Missing code or state", status_code=400)
user: KhojUser = await aget_user_by_uuid(state) # <-- No verification!
await NotionConfig.objects.filter(user=user).adelete() # <-- Deletes victim's config
# ... OAuth token exchange ...
access_token = final_response.get("access_token")
await NotionConfig.objects.acreate(token=access_token, user=user) # <-- Stores attacker's token
To exploit is relatively easy. Once we know the victim's UUID, we simply initiate the Notion sync process on our own account and intercept the callback, replacing the state parameter with the victim's UUID.
Impact
Deletes user's existing Notion sync and replaces it with attacker-controlled Notion. Could allow for index poisoning. I'm not entirely sure what Khoj does with synced files but if it's being passed as context to an LLM then I can imagine there's potential here.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐍PyPI | khoj | all versions | No fix |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for khoj. 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.
Remediation status
No patched version of khoj has shipped for GHSA-6whj-7qmg-86qj yet. Where your build allows, override or pin the dependency away from the vulnerable range, and apply any maintainer-recommended mitigation.
Mitigate without a patch
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-6whj-7qmg-86qj 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-6whj-7qmg-86qj. 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-6whj-7qmg-86qj in your dependencies?
O3 detects GHSA-6whj-7qmg-86qj across PyPI dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.