Authorization Bypass in I hate money
Impact
An authenticated member of one project can modify and delete members of another project, without knowledge of this other project's private code. This can be further exploited to access all bills of another project without knowledge of this other project's private code.
With the default configuration, anybody is allowed to create a new project. An attacker can create a new project and then use it to become authenticated and exploit this flaw. As such, the exposure is similar to an unauthenticated attack, because it is trivial to become authenticated.
Patches
ihatemoney/models.py | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/ihatemoney/models.py b/ihatemoney/models.py
index fe7b519..5691c75 100644
--- a/ihatemoney/models.py
+++ b/ihatemoney/models.py
@@ -380,7 +380,7 @@ class Person(db.Model):
def get_by_name(self, name, project):
return (
Person.query.filter(Person.name == name)
- .filter(Project.id == project.id)
+ .filter(Person.project_id == project.id)
.one()
)
@@ -389,7 +389,7 @@ class Person(db.Model):
project = g.project
return (
Person.query.filter(Person.id == id)
- .filter(Project.id == project.id)
+ .filter(Person.project_id == project.id)
.one()
)
Workarounds
To limit the impact, it is possible to disable public project creation by setting ALLOW_PUBLIC_PROJECT_CREATION = False in the configuration (see documentation). Existing users will still be able to exploit the flaw, but this will prevent an external attacker from creating a new project.
For more information
Person.query.get() and Person.query.get_by_name() were mistakenly running a database join on the Project table without constraining the result.
As a result, Person.query.get(42, "projectfoo") would return the Person with id=42, even if it is not associated to the project "projectfoo". The only condition is that "projectfoo" must exist.
This flaw can be exploited in several places:
-
API: PUT requests to /api/projects/<project>/members/<personID> will succeed even though <personID> is not a member of <project>.
This allows an authenticated attacker to alter the state of a member (name, weight, activated) in any project. In addition, the altered member will no longer be associated with its original project but will be associated to the attacker project instead, breaking many features of IHateMoney. For instance, bills referencing the altered member will no longer be visible in the original project.
This causes an additional information disclosure and loss of integrity on bills: the attacker will now be able to see, edit and delete bills belonging to the altered member, because IHateMoney now believes that these bills are associated to the attacker project through the altered member.
For instance, assume that Person(id=42) is a member of project "targetProject", and that the attacker has access to another project "attackerProject" with the private code "attackerPassword". The attacker can modify Person(id=42) with this command:
$ curl -X PUT -d "name=Pwn3d&activated=1" --basic -u attackerProject:attackerPassword http://$SERVER/api/projects/attackerProject/members/42
The attacker can now see, edit and delete bills paid by Person(id=42) by simply browsing to http://$SERVER/attackerProject/
-
Editing a member through the web interface at /<project>/members/<personID>/edit will succeed even though <personID> is not a member of <project>.
This is very similar to the PUT exploit. Reusing the same example, the attacker needs to login to its "attackerProject" project with the private code "attackerPassword". It can then alter the state of Person(id=42) by accessing the edit form at the following URL:
http://$SERVER/attackerProject/members/42/edit
Again, as a result of the alteration, the altered member will become associated to the project "attackerProject", resulting in the same information disclosure and loss of integrity on bills.
-
API: DELETE requests to /api/projects/<project>/members/<personID> will similarly allow to delete the member <personID> even if it belongs to a different project than <project>.
$ curl -X DELETE --basic -u attackerProject:attackerPassword http://$SERVER/api/projects/attackerProject/members/42
The impact is less serious than with PUT, because DELETE only deactivates a member (it does not really delete it).
All these exploits require authentication: an attacker needs to know a valid project name and its associated "private code". Once this requirement is fullfilled, the attacker can exploit this flaw to alter the state of members in any other project, without needing to know the target project name or its private code.
Person.query.get_by_name() suffers from the same issue as Person.query.get(). It has an additional issue: if multiple Person objects with the same name exist (this is possible if they are associated to different projects), get_by_name() will crash with MultipleResultsFound because of the call to one().
However, since Person.query.get_by_name() is currently not used anywhere in IHateMoney, the bug affecting this function has no impact and is not exploitable.
Latest trending attack
Criticalyesterday
Google Brotli decoder - Remote Code Execution via compressed requests (Exploitation Check)
This control is the second step of the Patrowl response to CVE-2026-111497 (Node2Shell). It complements the first control of 22 October 2026, which focused on the Node.js runtime, and does not replace it.
Why a second control
CVE-2026-111497 is an out-of-bounds write in the Google Brotli decoder, the library that decompresses Brotli ("br") data. It was first published as a Node.js vulnerability, because Node.js bundles this decoder. Later publications showed that the flaw is in the library itself. Every product that embeds an affected version of the decoder is concerned, whatever its language or vendor. A public proof of concept demonstrates code execution.
Any server that decompresses request bodies sent with "Content-Encoding: br" exposes the decoder to unauthenticated attackers. Known cases include Express 5 and NestJS 11 on Node.js, Koa and Fastify setups, Bun, Envoy and Istio with request decompression enabled, and any application or reverse proxy that decodes Brotli request bodies with an affected library.
The flaw also works in the other direction. A malicious server can compromise an unpatched client that decodes a Brotli response, for example Node.js fetch(), Python clients (requests, httpx, aiohttp, Scrapy) with the Brotli or brotlicffi package, or libcurl builds with Brotli support.
What this control checks
Patrowl sends small, valid Brotli-compressed requests to each asset and observes whether the server decodes them. The check does not depend on the technology: it detects the behaviour, whatever the stack behind it. These requests are what any legitimate client could send. They never reach the vulnerable code path and cannot affect the availability of the service.
How to read the result
Potentially impacted means the asset decodes Brotli request bodies: the attack surface is exposed. It does not confirm that the asset is vulnerable. The version of the Brotli decoder in use is not visible from the outside.
This check covers applications that parse request bodies on their root path. A Not impacted result does not rule out exposure on other routes, nor on the client side. Checking the versions on the servers remains the only exhaustive verification.
Affected and fixed versions
Google Brotli 1.0.3 to 1.2.0 are affected, fixed in 1.0.10, 1.1.1 and 1.2.1. Fixed releases of the main products embedding it: Node.js 22.23.4, 24.22.1 and 26.12.1 (end-of-life lines 20.x and older and 25.x receive no fix), Bun 1.4.4, Envoy 1.37.8, 1.38.6, 1.39.3 and 1.40.0, Istio 1.29.9, 1.30.6 and 1.31.2, Python Brotli 1.2.1 and brotlicffi 1.2.1.0. Products that link against the system library depend on the Brotli package of the Linux distribution.
Recommendations
Update every component that embeds the Brotli decoder (only real fix). Start with the assets flagged by this control. Restart the processes after the update. Rebuild containers, single-executable applications and appliances that bundle an affected runtime or library.
Reduce the exposure until the update is deployed. Reject requests with "Content-Encoding: br" at the reverse proxy, load balancer or WAF, case-insensitively and including lists such as "gzip, br". On Envoy and Istio, disable request decompression for Brotli.
Review outbound calls. List the features that fetch content from servers you do not control, and patch the services that host them first.