[fix] Key the lifecycle capability table by the wire harness spellings - #6364
Open
mmabrouk wants to merge 1 commit into
Open
[fix] Key the lifecycle capability table by the wire harness spellings#6364mmabrouk wants to merge 1 commit into
mmabrouk wants to merge 1 commit into
Conversation
harnessKind matched only the bare literals 'pi', 'claude', 'codex', but the wire carries 'pi_core' / 'pi_agenta' (empty defaults to pi_core), so every playground Pi run resolved to 'unknown' and took the fail-closed capability row where every facet rebuilds the sandbox. Two consequences: - The one live route that already ships (a model-only switch applies live via setModel) never fired for Pi: every model switch in the playground tore down the warm sandbox, minted fresh Daytona Secrets, and re-entered the placeholder-propagation race (see #6362). - Every reconcile shadow log line for Pi read 'harness=unknown' with an all-rebuild plan, so the step-4/5 DISAGREE audit was noise for the main harness. Normalize exactly as buildRunPlan does before keying the capability table. Claude-Session: https://claude.ai/code/session_014s6jqKCsVKsNMmnrJVJkZt
dosubot
Bot
added
the
size:S
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Contributor
📝 WalkthroughSummary by CodeRabbit
WalkthroughThe runner now maps absent, ChangesPi harness routing
Estimated code review effort: 2 (Simple) | ~10 minutes Merge Risk: 🔵 Low · up to The change lets supported Pi requests reuse warm sandboxes for model-only switches, reducing unnecessary rebuilds. Malformed harness values can still select the live Pi path instead of failing closed, and equivalent omitted and explicit pi_core values may trigger unnecessary rebuilds; the PR is mergeable with explicit owner awareness and follow-up on validation and canonicalization. 🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches 💡 1📝 Generate docstrings 💡
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository YAML (base), Organization UI (inherited)
Review profile: CHILL
Plan: Pro Plus
Run ID: a0b663a4-0e97-482b-97ed-b38f8538c2d6
📒 Files selected for processing (2)
services/runner/src/lifecycle/reconciliation-router.tsservices/runner/tests/unit/lifecycle-reconcile-plan.test.ts
Included review availability: Your plan provides up to 8 included reviews per hour; 5 remain after this review.
This was referenced Aug 29, 2026
mmabrouk
added a commit
that referenced
this pull request
Aug 30, 2026
Found by a code audit of the cold/warm flows (follow-up to #6364). The coordinator computed the eviction reason with an else-if chain and the live model route then set mismatch = undefined wholesale, so a repair of the CONFIG question also cleared every question the chain never reached: - a model switch riding an EDITED transcript continued warm on a native conversation still holding the unedited turn; - riding a ROTATED credential with no delivery port, it ran on the old baked key (silently, if the old key still worked); - riding an EXPIRING mount lease, it let the turn die under the mount; - riding a stale tail, it ran a turn the tail check should have evicted. The claude live route has been exposed since step 6 shipped; #6364 makes the route fire for pi_core too, widening the surface, so this belongs in the same stack. Each repair door now marks ITS reason repaired and re-asks the ordered checks. A model switch with a deliverable rotation chains both repairs and stays warm (pinned); the four bad pairings rebuild (pinned). Claude-Session: https://claude.ai/code/session_014s6jqKCsVKsNMmnrJVJkZt
mmabrouk
added a commit
that referenced
this pull request
Aug 30, 2026
Found by a code audit of the cold/warm flows (follow-up to #6364). The coordinator computed the eviction reason with an else-if chain and the live model route then set mismatch = undefined wholesale, so a repair of the CONFIG question also cleared every question the chain never reached: - a model switch riding an EDITED transcript continued warm on a native conversation still holding the unedited turn; - riding a ROTATED credential with no delivery port, it ran on the old baked key (silently, if the old key still worked); - riding an EXPIRING mount lease, it let the turn die under the mount; - riding a stale tail, it ran a turn the tail check should have evicted. The claude live route has been exposed since step 6 shipped; #6364 makes the route fire for pi_core too, widening the surface, so this belongs in the same stack. Each repair door now marks ITS reason repaired and re-asks the ordered checks. A model switch with a deliverable rotation chains both repairs and stays warm (pinned); the four bad pairings rebuild (pinned). Claude-Session: https://claude.ai/code/session_014s6jqKCsVKsNMmnrJVJkZt
mmabrouk
added a commit
that referenced
this pull request
Aug 30, 2026
Cold/warm audit finding 2. modelCapabilities is the resolved model's input modalities: per-turn data the attachment chain reads from each incoming request, baked into nothing. But it rode the fingerprint and the harnessSession facet, and it changes WITH the model — so switching between a vision model and a text-only model moved two facets, the mixed plan refused the live setModel route, and the switch rebuilt the warm sandbox (fresh Daytona Secrets, the substitution race, ~15s) for nothing. The Pi catalog splits 249 text+image vs 163 text-only models, so this hit ordinary playground use even after #6364. Same treatment as workflowRevision/isDraft in step 1: out of the hash, documented in the volatile list, pinned in both directions (the per-turn-volatile test gains the field; the old changes-the-fingerprint test is reversed with the reason). Claude-Session: https://claude.ai/code/session_014s6jqKCsVKsNMmnrJVJkZt
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Context
While tracing why a config change evicted a user's warm sandbox (the placeholder-401 investigation, #6362), the reconcile shadow log showed
harness=unknownon every playground run. The lifecycle router'sharnessKindmatched only the bare literalspi,claude,codex, but the wire carriespi_coreorpi_agenta, and an empty harness defaults topi_core. Every playground Pi run therefore resolved tounknownand took the fail-closed capability row, where every facet maps to a full sandbox rebuild.Two real costs:
setModel) never fired for Pi. Every model switch in the playground tore down the warm sandbox, created fresh Daytona Secrets, and re-entered the substitution-propagation race that produces the placeholder 401s.Changes
harnessKindnow normalizes exactly asbuildRunPlandoes before keying the capability table.Before:
harness: "pi_core"resolved tounknown(model switch plansrebuild-sandbox).After:
pi_core,pi_agenta, barepi, and the empty default all resolve topi(model switch plansapply-live; a harness-session change plansreopen-session).claudeandcodexwere already correct and are unchanged. A genuinely unknown harness still fails closed.Tests
lifecycle-reconcile-plan.test.tspin the wire spellings, the empty default, and the recoveredapply-livemodel route.pnpm run typecheckclean.What to QA
https://claude.ai/code/session_014s6jqKCsVKsNMmnrJVJkZt