ADR 0027: residual-risk wording understates the dir= filing boundary (host-scoped, not clone-scoped) #263
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#263
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 follow-up.
docs/adr/0027L131-132 describes the report carve-out's residual risk as filing junk "to trackers of repos the operator has cloned and authed". The actual boundary is wider: the clone-on-disk is fabricable.git initplusgit remote add origin <any repo on an authed forge>makes ANY repo on any forge with a host-keyed credential a valid filing target -newSvcForDirresolves owner/repo from the fabricated origin andGetConfigForDirsupplies the host-keyed token.This is consistent with the declared accident-detector posture (the agent has shell and the unguarded CLI anyway), so no behavior change is required - but the ADR's residual-risk section should state the boundary honestly: host-scoped by credential, not clone-scoped, with the on-disk clone acting only as friction. The deferred
requiresUserInteractionbelt mentioned in the ADR remains the real fix if a harder boundary is ever wanted.Acceptance: ADR 0027 residual-risk paragraph amended to describe the host-scoped boundary and the fabricable-clone caveat.