drift.yml: socket-privileged forgejo-smoke job is reachable from pull_request events #265
Labels
No labels
breaking
bug
documentation
enhancement
epic
good first issue
help wanted
refactoring
resolution/duplicate
resolution/invalid
resolution/wontfix
security
severity/critical
severity/high
severity/low
severity/medium
status/abandoned
status/blocked
status/needs-decision
status/needs-info
status/needs-verification
testing
upstream
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
sixfold-space/madtea#265
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Security audit, highest-priority CI finding. The drift workflow triggers on
pull_request(.forgejo/workflows/drift.yml:36), and its forgejo-smoke job - gated only byvars.RUN_FORGEJO_SMOKE == 'true'(drift.yml:99) - usesdocker execagainst the HOST docker socket to provision the sidecar Forgejo admin (drift.yml:167) and then executes PR-modifiable content (tests/integration/forgejo-smoke.sh, plus the run blocks themselves, drift.yml:183).Consequence: on any runner where that variable is set and the docker socket is exposed, a malicious fork PR that edits the workflow or the smoke script gets code execution with host-docker-socket access, which is effectively host root. On Codeberg the vari
Duplicate of #266, which covers the same drift.yml pull_request + host-docker-socket pwn-request finding with the full two-option fix (confirm the org runner requires fork-PR approval, or drop pull_request from the socket-privileged smoke job) and severity labels. Consolidating there.