MaxKB code injection in custom tool validator rules
Affects MaxKB 2.10.6-lts and earlier, where a logged-in workspace account can create a custom tool with a crafted Init Param validator rule. Fixed in 2.10.7-lts.
- Exploited
- No known exploitation
- Fix
- Fixed in 2.10.7-lts
- Who is exposed
- Specific setups only
Summary
MaxKB by 1Panel-dev has a flaw in how custom tool Init Param validator rules are handled. A logged-in user with low privileges can store JavaScript that runs in the MaxKB web origin when the tool's Debug screen is opened, and the attack is reached over the network but needs a user to take action such as opening that screen.
Written by a model
Original description
MaxKB is an open-source AI assistant for enterprise. Prior to 2.10.7-lts, custom Tool Init Param definitions accept attacker-controlled props_info.rules[].validator strings through POST /admin/api/workspace/{workspace_id}/tool and PUT /admin/api/workspace/{workspace_id}/tool/{tool_id}, store them without restriction, and pass them to eval() when the parameter form renders. An authenticated user with the default USER workspace role can store JavaScript and trigger execution by opening the tool's Debug screen before entering or submitting a parameter value. The same Init Param form is exposed to accounts with edit permission on the tool, but execution by a different account was not independently verified. The confirmed script runs in the MaxKB origin; access to another user's bearer token and authenticated account authority is plausible only if that user renders the malicious form. This issue is fixed in version 2.10.7-lts.
Who should check
Teams running MaxKB 2.10.6-lts or earlier that use custom tools with Init Param fields, including workspace members with edit access to those tools
“Any authenticated user — including an account holding only the default, lowest-privilege workspace role — can define a tool parameter whose validation rule is a raw JavaScript string.”
What to do
Check whether your setup matches what the bug needs.
MaxKB 2.10.6-lts and earlier (affected versions are <=v2.10.6-lts) Attacker needs a logged-in account; the default lowest-privilege workspace role is enough to create the malicious tool The injected script runs when the tool's Init Param form is rendered (confirmed when the creating account opens its own Debug screen)Upgrade to a fixed version: 2.10.7-lts.
Check your systems with an agent
Paste this into a coding or ops agent. It checks your systems read-only and reports a table per asset.
Check whether my systems are affected by CVE-2026-129998 (MaxKB: Eval Injection via Custom Tool "Init Param" Validator Rule Allows Arbitrary JavaScript Execution), using the cve.now data below.
Rules: work read-only. Do not run exploit code, proof-of-concept scripts or scanners that send attack payloads, and do not change any configuration, package, image or deployment. Report findings only.
What cve.now has for CVE-2026-129998 (entry last updated 2026-11-20):
Affected:
- 1Panel-dev MaxKB: fixed in 2.10.7-lts
Fixed versions: 2.10.7-lts.
Conditions (all must be true for the bug to be hit):
- MaxKB 2.10.6-lts and earlier (affected versions are <=v2.10.6-lts) (source: https://github.com/1Panel-dev/MaxKB/security/advisories/GHSA-c4xr-7rr6-43mr)
- Attacker needs a logged-in account; the default lowest-privilege workspace role is enough to create the malicious tool (source: https://github.com/1Panel-dev/MaxKB/security/advisories/GHSA-c4xr-7rr6-43mr)
- The injected script runs when the tool's Init Param form is rendered (confirmed when the creating account opens its own Debug screen) (source: https://github.com/1Panel-dev/MaxKB/security/advisories/GHSA-c4xr-7rr6-43mr)
Setup: unknown whether the bug needs a non-default feature or setting. Check the advisory.
Attacker needs: network access, low-privilege login, a user action.
CISA KEV: CVE-2026-129998 is not listed as of cve.now's last read.
This KEV status is for CVE-2026-129998 only. If you also check another CVE, look up its own KEV status; never assume it from this one.
How to check:
- Installed software: check what is actually running, not what a repo declares. Read the version from each host, container image and Kubernetes node (the binary's version output, the OS package database, image layers), and check whether the conditions above hold there. Distribution packages may carry a backported fix under an older version number; confirm against the distribution's own advisory.
Verify against the linked advisories before concluding, and cite the date you read each one:
- https://github.com/1Panel-dev/MaxKB/security/advisories/GHSA-c4xr-7rr6-43mr
- https://github.com/1Panel-dev/MaxKB/pull/7257
- https://github.com/1Panel-dev/MaxKB/commit/fad2427ab46558c5fad764993a08e3661ab4b7f9
- https://github.com/1Panel-dev/MaxKB/releases/tag/v2.10.7-lts
Do not conclude an asset is unaffected unless you checked it and the evidence covers both its version and the conditions above. If you could not check something, mark it unknown and say what is missing.
Report a table with one row per asset and these columns: Asset | Version or build | Status (affected, fixed, not applicable, unknown) | Compensating controls | Exposure (internet-facing, internal, none) | Evidence | Next action.
Entry: https://cve.now/CVE-2026-129998
For follow-up questions, use the cve.now MCP server at https://mcp.cve.now/mcp (no authentication), for example execute with name cve.get and arguments {"id": "CVE-2026-129998"}.Affected versions
- 1Panel-dev MaxKBfixed in 2.10.7-lts
cve.now Index
- Exploitation?/40Unknown. No CISA SSVC exploitation rating yet
- Exposure8/25Specific setups only, reachable over the network
- Reach?/20Unknown. No tracked technology or package matched
- Severity7.3/10CVSS 7.3
- No patch0/5No signal
What would change the score
- Would score about 65 if added to CISA KEV
- Would score about 60 if CISA's SSVC rates exploitation active
- Would score about 22 if CISA's SSVC rates it automatable
Timeline
- CVE published
- First seen by cve.now
Evidence 6
| Kind | Source |
|---|---|
| Vendor advisory 1Establishes: Affected and fixed versions. | |
| Advisory | GitHub · 1Panel-dev/MaxKB · Advisory GHSA-c4xr-7rr6-43mr |
| CVE record 2Establishes: The identifier and description. | |
| Record | CVE List · CVE-2026-129998 |
| Record | NVD · CVE-2026-129998 |
| Other links 3 | |
| Repository | GitHub · 1Panel-dev/MaxKB · Pull request #7257 |
| Repository | GitHub · 1Panel-dev/MaxKB · Commit fad2427a |
| Repository | GitHub · 1Panel-dev/MaxKB · Release v2.10.7-lts |
Questions about CVE-2026-129998
Is CVE-2026-129998 being exploited?
It is not in CISA's Known Exploited Vulnerabilities catalog, and CISA's SSVC assessment does not mark exploitation as active. That does not mean it has not been attacked: vendors and researchers can report attacks before CISA lists them. No public exploit code is tracked.
Source: cve.now feed data
What does this mean?
The validator rule stored for a custom tool's Init Param is passed directly to JavaScript's eval() and runs as soon as the tool's Init Param form is rendered, executing script in the MaxKB origin. The impact is equivalent to cross-site scripting, and session or bearer token theft is plausible but was not demonstrated.
Source: github.com
Which versions are affected by CVE-2026-129998?
MaxKB: before 2.10.7-lts.
Source: cve.now feed data
Am I affected, and how do I check?
MaxKB 2.10.6-lts and earlier are affected. Exposure requires a logged-in account, and the default lowest-privilege workspace role is enough to create the malicious tool. The injected script runs when the tool's Init Param form is rendered, which the advisory confirmed when the creating account opens its own Debug screen.
Source: github.com
How do I fix CVE-2026-129998?
Upgrade MaxKB to 2.10.7-lts or later, which is the listed fixed version.
Source: github.com
Related
- CVE-2026-77521MaxKB: Prompt-injectable agent can lead to command execution
- CVE-2026-79916MaxKB AWS Bedrock model credential injection leads to remote code execution
- CVE-2026-54149MaxKB MCP tool import validation bypass allows post-authentication remote code execution
- CVE-2026-44847MaxKB: Webhook Trigger Authentication Bypass
- CVE-2026-77523MaxKB: Cross-workspace model parameter form write
- CVE-2026-79917MaxKB: Chat share-link endpoint missing owner check: a chat token can publish another user's conversation
No corrections so far. Something wrong on this page? Report an error.