ledger: P1.3.3 closeout — challenge canon-parity landed (incl. process failure post-mortem); supersede predecessor handoff

[klappy]

Closeout ledger for P1.3.3 + flip predecessor handoff to superseded.

P1.3.3 is closed. Two halves, both load-bearing:

What shipped (technical)

• D5 stemmed prereq matcher — replaced per-prereq regex with stemmed set intersection + 4 structural-test side-paths (URL, numeric, proper-noun, citation). PRD /home/claude/work/prd-p1-3-3.md D3/D5.
• D9 cache removal — dropped cachedChallengeTypeIndex and inlined buildBM25Index per cache-fetches-and-parses. Cache existed for microsecond optimization on a per-request hot path that was already O(prereq_count × keyword_count).
• cache-fetches-and-parses graduated tier-2 canon (klappy.dev#125, merged 3726073) on third deciding-argument recurrence (P1.3.1 implicit, P1.3.2 explicit, P1.3.3 explicit).

All landed in oddkit 0.21.0 (PR #120 merged 33ca5bf, PR #121 merged 25ad719).

What failed (process)

• Orchestrator merged PRs odd/gate: canon governance docs for P1.3.2 gate vodka refactor (Phase 1) #120 and handoff: P1.3.2 Phase 2 forward handoff #121 with Cursor Bugbot still in_progress, treating it as non-blocking
• Skipped Sonnet 4.6 validator dispatch despite the P1.3.2 ledger explicitly warning that smoke-only should not become the default
• Bugbot subsequently posted 2 findings:
  • Medium — stop-word filter dropped from from source-named vocab on both sides of tokenize(), breaking the strictly-additive invariant the PR description claimed
  • Low — BasePrerequisite re-listed PrereqMatchVocab fields manually instead of using & PrereqMatchVocab intersection (DRY violation in the very interface introduced to be DRY)
• Approximate prod-regression window: ~1h 39m on from-keyword source-named matches (04:11Z → 05:09Z)

What got fixed (structural)

Three canon artifacts in klappy.dev#126 (merged ee9aee4):

1. NEW canon/constraints/release-validation-gate.md (tier 1, 168 lines) — three binding rules (no merge with active reviews, no promotion without independent fresh-context validation when load-bearing, canon outranks session-scoped recommendations)
2. NEW canon/principles/contract-governs-handoff-drift.md (tier 2, 128 lines) — graduated on third deciding-argument recurrence; the third recurrence was costed (principle's absence demonstrated to break things)
3. AMEND canon/bootstrap/model-operating-contract.md (+4 lines) — discoverability hook so future sessions encounter the rules on first turn

Captures O-open P11: oddkit_gate mechanical enforcement of release-validation-gate at execution → completion transitions.

Fix-forward applied the new canon end-to-end

Stage	Result
oddkit#122 (0.21.1 fix branch)	merged d17bc0c after Bugbot completed/success
Sonnet 4.6 validator (sesn_011CaERPjHi1CV4TrvKW68jB)	CONDITIONAL PASS (conditional only on canon merging first, resolved before promotion)
oddkit#123 (promotion)	Bugbot wait ~225s respected; completed/success; merged 2c5d652b
Prod warmup + smoke 3×	172/172 each
Live self-call	from fix confirmed in prod

Meta-instructive moment

Bugbot caught a within-canon DRY violation on the bootstrap hook in PR #126 itself. The first application of release-validation-gate's discipline caught me in another canon violation (dry-canon-says-it-once) before I could merge it. Working as designed.

Files changed

• NEW odd/ledger/2026-04-20-p1-3-3-challenge-canon-parity-landed.md (230 lines)
• AMEND odd/handoffs/2026-04-20-p1-3-3-challenge-revisit.md (+1, -1 line) — frontmatter status: active → status: superseded with superseded_by and supersession_note naming the "Option A is fine for P1.3.3" line as the recommendation that produced the incident

Carry-forward O-opens

ID	Status	Description
P2/P3/P5/P6/P8/P9/P10	carried	unchanged from prior sweeps
P11	NEW	oddkit_gate mechanical enforcement of release-validation-gate at execution → completion
P12	NEW	tokenize() audit pass — explicit stop-word set at every canon-vocab/canon-input call site

Refs

• Closeout ledger: this PR's odd/ledger/2026-04-20-p1-3-3-challenge-canon-parity-landed.md
• Validator report: /home/claude/work/validator-reports/p1-3-3-validator-report.md
• Encoded artifact: /home/claude/work/encodes/p1-3-3-structural-fix.md
• Canon: klappy://canon/constraints/release-validation-gate, klappy://canon/principles/contract-governs-handoff-drift, klappy://canon/principles/cache-fetches-and-parses
• Code: oddkit#120 (33ca5bf), handoff: P1.3.2 Phase 2 forward handoff #121 (25ad719), odd/gate/transitions: add inflected forms for geminating verbs #122 (d17bc0c), odd: P1.3.2 closeout ledger + supersede obsolete handoffs #123 (2c5d652b)
• Canon PRs: klappy.dev#125 (3726073), canon: release-validation-gate + contract-governs-handoff-drift + bootstrap hook (P1.3.3 root-cause fix) #126 (ee9aee4)

---

Note

Low Risk
Low risk: this PR only updates markdown handoff/ledger documentation and does not change runtime code or operational configuration.

Overview
Adds a new closeout ledger odd/ledger/2026-04-20-p1-3-3-challenge-canon-parity-landed.md documenting the P1.3.3 outcomes (what shipped, what failed, and the structural process fixes applied).

Updates the prior handoff odd/handoffs/2026-04-20-p1-3-3-challenge-revisit.md to status: superseded, linking to the closeout ledger and warning future sessions not to follow the handoff’s validation-plan recommendation.

^{Reviewed by Cursor Bugbot for commit fb14edb. Bugbot is set up for automated code reviews on this repo. Configure here.}