CedarCopilot wants to merge 0 commits into staging from chore/aspire-playbook-toolchain-and-audit-docs
Live on prod, no production signal yet
Strict CloudWatch fallback saw 169 success-shaped log lines matching pipeline, analytics, quota, aop, but no tied operation was present, so Libra is not calling this working.
Strict CloudWatch fallback found 1 recent prod failure log lines for [UpdateExternalCrmWorkflow] Driver update failed (Step updateDeal failed: HTTP 400: {\), but the same failure had 8 log lines in the comparable pre-deploy baseline. Libra is not blaming this PR intent without.
Strict CloudWatch fallback saw 513 success-shaped log lines matching pipeline, analytics, quota, aop, but no tied operation was present, so Libra is not calling this working.
Libra has verdicts on 0 of 3 tracked behaviors on prod; 3 are still being checked. Libra checks hourly for 3 days after each deploy.
Preserves the AspireIQ playbook-build work and captures the reusable method for future audits. Grounded in org 2d2b90fb (AspireIQ). No changes to product/runtime code.
write-agentic was dropping objection.response / use_case / feedback from CANONICAL_FIELDS; restored.aspire-persona-rollup.ts), apply-curation.py, inject-taxonomy-into-playbook.py, aspire-canon-pipeline.sh, run wrapper, aspire-playbook-rendercheck.mjs (render-smoke harness), + write-judgment.ts PG_ONLY/scoped restore.resources/playbook-build-runbook.md (org-agnostic build manual) + PROCESS-DELTAS.md (8 method deltas vs the canonical skill, staged for reconciliation); SKILL.md points to both.docs/audits/aspire/ (funnel-per-deal, themes, use-case defs + quote-excludes, pricing groups, per-persona curation, vocab canon/seed) + the handoff doc.docs/audits/aspire/artifacts/ (playbook SPA + data-lineage); requirements doc renamed → aspireiq-audit-handoff.md (the account's single source of truth).Regenerable rollup dumps, candidate pools, and the 15 MB judgment-backfill/ (data is back in the DB) , all rebuild via the toolchain.
PROCESS-DELTAS.md)chapter-format/build.py deliverable path , sanction as a new template vs migrate.deal-intelligence-critique pass hasn't been run over the playbook.🤖 Generated with Claude Code
https://claude.ai/code/session_01M6QhCHuWDjeRtdNArHoLEg
The PR adds an AspireIQ-specific taxonomy-to-playbook toolchain, reusable audit process documentation, curated inputs, and generated HTML deliverables. It also expands taxonomy writer coverage and adds a Postgres-only judgment restore mode.
The PR should not merge until the one-shot pipeline propagates failures and the PG_ONLY restore can no longer leave searchable judgment data stale.
The documented build command currently reports success despite failed or absent stages, while the new Postgres-only restore creates a split state between Postgres and Turbopuffer that existing intelligence searches can observe.
Files Needing Attention: apps/server/scripts/aspire-canon-pipeline.sh, apps/server/scripts/write-judgment.ts
| Filename | Overview |
|---|---|
| apps/server/scripts/aspire-canon-pipeline.sh | Adds the documented one-shot orchestration, but suppresses every child failure and references two absent wrappers. |
| apps/server/scripts/write-judgment.ts | Adds a Postgres-only restore path that temporarily leaves searchable Turbopuffer judgment data stale. |
| apps/server/scripts/write-agentic.ts | Replaces a hand-maintained field allowlist with registry-derived routing and rejects unknown assembled fields. |
| apps/server/scripts/aspire-persona-rollup.ts | Adds the deterministic AspireIQ aggregation pipeline; several data assumptions remain operationally specific and it violates the absolute-import convention. |
| apps/server/scripts/inject-taxonomy-into-playbook.py | Adds deterministic taxonomy and curat |
Libra has not measured any production surfaces for this change yet.