chore: planner run 2026-03-21

planner/MEMORY.md

@@ -0,0 +1,26 @@
+# Planner Memory
+
+## 2026-03-21 — Second planner run
+
+### Milestone state
+- **Foundation**: COMPLETE. Agent loop, supervisor, planner, multi-project all working.
+- **Adoption**: IN PROGRESS. Bootstrap (#393), docs (#394), dashboard (#395), example project (#466) all in backlog. No work started yet — issues were unlabeled until this run fixed them.
+- **Scale**: PARTIALLY started (multi-project works for 3 repos).
+
+### Patterns
+- **Predictor over-signals on transient ops issues**: 4/6 predictions this run were orphaned tmux sessions or crashed reviews — things the supervisor handles automatically. Expected to continue until predictor learns to filter supervisor-handled issues.
+- **Label loss on issue creation**: The 3 Adoption issues created last run had no labels when checked this run. Root cause unknown — could be a silent API failure in the planner's issue creation, or labels removed by another process. Watch for recurrence.
+- **Long tech-debt backlog blocks features**: ~20 small backlog items (tech-debt, bug fixes) will be processed before Adoption features due to sequential pipeline and lower issue numbers. Not a problem per se — maintains factory health — but means Adoption work won't start for weeks unless manually prioritized.
+- **needs_human is a silent pipeline killer**: When a dev-agent writes PHASE:needs_human and no human responds, the pipeline stalls silently. Supervisor doesn't escalate. Filed #465 to fix.
+
+### Strategic direction
+- Adoption remains the leverage multiplier. All 4 Adoption issues are now in backlog: #393 (init) → #394 (docs) → #395 (dashboard), plus #466 (example project, depends on #393).
+- The critical path is: #393 (init) must land first — docs and example project both reference it.
+- #465 (supervisor needs_human escalation) is operational leverage — prevents the kind of silent stall observed via #446.
+
+### Watch list
+- Label persistence: verify #393/#394/#395 retain their backlog labels next run
+- Tech-debt throughput: how fast is the dev-agent clearing the backlog queue?
+- #357 (in-progress): action-agent runtime isolation — track completion
+- #448 (prediction/backlog): disk at 75%, trend improving
+- #446 (prediction/backlog): harb needs_human pattern — is #465 picked up?

planner/journal/2026-03-21.md

@@ -0,0 +1,32 @@
+# Planner run — 2026-03-21 09:29 UTC
+
+## Predictions triaged
+- #455: DISMISS — orphaned gardener session, transient, supervisor's job
+- #454: DISMISS — crashed review session, transient, supervisor recovers
+- #449: DISMISS — legacy predictor duplication, already tracked by #419
+- #448: WATCH — disk at 75% (was 79% last run), improving trend, supervisor monitors
+- #447: DISMISS — swap at 52%, expected behavior with memory guards
+- #446: WATCH — harb pipeline stalled 8h on needs_human, supervisor didn't escalate
+
+## Issues created
+- #465: feat: supervisor escalates prolonged PHASE:needs_human states — revealed by #446 pattern, prevents silent pipeline stalls
+- #466: feat: example project demonstrating the full Disinto lifecycle — VISION.md Adoption gap, needed by docs and landing page
+
+## Label fixes
+- #393 (disinto init): added backlog label — was created last run but lost its label
+- #394 (quickstart docs): added backlog label — same issue
+- #395 (metrics dashboard): added backlog label — same issue
+
+## Observations
+- Predictor continues to over-signal on transient operational issues (4/6 predictions were transient tmux/session issues the supervisor already handles). Pattern from last run confirmed.
+- Adoption issues from last planner run (#393/#394/#395) existed but had NO labels. The dev-agent only picks up backlog-labeled issues. Root cause unclear — either the label API call failed silently during creation, or labels were removed. Fixed this run.
+- Foundation milestone remains complete. Adoption still the bottleneck — no progress since last run because issues weren't in the backlog.
+- Tech-debt and small backlog items (~20) will be processed before Adoption features due to sequential pipeline and lower issue numbers.
+- #357 (action-agent runtime isolation) is in-progress — active dev work happening.
+
+## Deferred
+- Scale milestone (multi-project works, plugin system premature without users)
+- Evidence pipeline (harb-specific, blocked on #1047)
+- Production halt/resume (#1) — far future, no users to protect yet
+- Multi-VPS (#4) — premature, single server handles current load
+- Adding backlog labels to #462 (PHASE:escalate) and #291 (secrets in formulas) — both valid but not highest leverage this cycle