change-interpreter cannot read tickets from worktree subagents (no tracker configured) #498
Labels
No labels
P0
P1
P2
P3
bug
create
delete
enhancement
filed-by/agent
filed-by/user
frozen
lint-rule
needs-info
needs-triage
next
plugin/cc-architect
plugin/os
plugin/os-adr
plugin/os-aidd-lint
plugin/os-backlog
plugin/os-context
plugin/os-doc-hygiene
plugin/os-sdlc
plugin/os-vault
project/cc-os
ready-for-agent
ready-for-human
recurring
review
update
waiting
wayfinder:grilling
wayfinder:map
wayfinder:map
wayfinder:research
wayfinder:task
wayfinder:task
wontfix
worklist/deviations
worklist/lint-rule
worklist/new-implement-build
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set
Reference
jared/cc-os#498
Loading…
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?
Context
2026-08-24 #491 run, implementation 2: change-interpreter settled verdict:error reporting 'no tracker configured', despite a valid .cc-os symlink in the worktree; the main session read the ticket fine. Cost: one wasted implementation-open. Workaround that worked: inline the ticket body in the dispatch brief.
Task
Either have the runner inline the ticket body into the change-interpreter brief at implementation-open, or diagnose why ConfigLocator/tea fails in the subagent environment. Evidence: .sdlc/pipeline.db implementation 2, handoff 1.
Origin
Resolution
Done: Root cause found and fixed via #500 (merged
a0b969c, ADR-0146). The diagnosis in this ticket was misdirected: tracker resolution never failed in the subagent env — it never ran anywhere. RoundCommands.subagent_start (round_commands.rb:22) passed no tracker, so StartSubagent unconditionally defaulted to NullTracker; TeaTracker.for had zero production call sites. The failure was identical from the main checkout. #500 implemented this ticket's first proposed option: body fetched once at implementation-open (orchestrator cwd), persisted in implementations.ticket_body, read into the brief by BriefFacts; NullTracker#fetch_body now raises TrackerUnavailable instead of emitting the placeholder.Evidence: plugins/os-sdlc/tests/runner/cli_subagent_start_test.rb test_entry_agents_brief_carries_the_ticket_body_with_a_null_handoff (green); open_implementation_ticket_body_test.rb; commit
b25598eon main. Live end-to-end confirmation will come free with the planned #492 replay — if its change-interpreter receives a real brief from a worktree, this is proven in production.Follow-ups: none — the #492 replay doubles as live verification; if it reproduces a placeholder brief, reopen this.