CVE-2023-38487

HedgeDoc is software for creating real-time collaborative markdown notes. Prior to version 1.9.9, the API of HedgeDoc 1 can be used to create notes with an alias matching the ID of existing notes. The affected existing note can then not be accessed anymore and is effectively hidden by the new one.

When the freeURL feature is enabled (by setting the allowFreeURL config option or the CMD_ALLOW_FREEURL environment variable to true), any user with the appropriate permissions can create a note by making a POST request to the /new/<ALIAS> API endpoint. The <ALIAS> parameter can be set to the ID of an existing note. HedgeDoc did not verify whether the provided <ALIAS> value corresponds to a valid ID of an existing note and always allowed creation of the new note. When a visitor tried to access the existing note, HedgeDoc will first search for a note with a matching alias before it searches using the ID, therefore only the new note can be accessed.

Depending on the permission settings of the HedgeDoc instance, the issue can be exploited only by logged-in users or by all (including non-logged-in) users. The exploit requires knowledge of the ID of the target note. Attackers could use this issue to present a manipulated copy of the original note to the user, e.g. by replacing the links with malicious ones. Attackers can also use this issue to prevent access to the original note, causing a denial of service. No data is lost, as the original content of the affected notes is still present in the database.

This issue was fixed in version 1.9.9. As a workaround, disabling freeURL mode prevents the exploitation of this issue. The impact can be limited by restricting freeURL note creation to trusted, logged-in users by enabling requireFreeURLAuthentication/CMD_REQUIRE_FREEURL_AUTHENTICATION.

Weakness

The product performs authentication based on the name of a resource being accessed, or the name of the actor performing the access, but it does not properly check all possible names for that resource or actor.

Affected Software

Potential Mitigations

• Assume all input is malicious. Use an “accept known good” input validation strategy, i.e., use a list of acceptable inputs that strictly conform to specifications. Reject any input that does not strictly conform to specifications, or transform it into something that does.
• When performing input validation, consider all potentially relevant properties, including length, type of input, the full range of acceptable values, missing or extra inputs, syntax, consistency across related fields, and conformance to business rules. As an example of business rule logic, “boat” may be syntactically valid because it only contains alphanumeric characters, but it is not valid if the input is only expected to contain colors such as “red” or “blue.”
• Do not rely exclusively on looking for malicious or malformed inputs. This is likely to miss at least one undesirable input, especially if the code’s environment changes. This can give attackers enough room to bypass the intended validation. However, denylists can be useful for detecting potential attacks or determining which inputs are so malformed that they should be rejected outright.

References

• https://github.com/hedgedoc/hedgedoc/pull/4476/commits/781263ab84255885e1fe60c7e92e2f8d611664d2
• https://github.com/hedgedoc/hedgedoc/security/advisories/GHSA-7494-7hcf-vxpg