- Add import_cucloud_pricing.go for 联通云 payg 公开价抓取
- Add import_bytedance_pricing.go for 火山引擎/ByteDance Ark 定价导入
- Include test files and sample testdata for both importers
- Update plan catalog inventory docs and seeds
- Add cucloud pricing importer implementation plan
- Align pipeline scripts and smoke gate tests
- official_pricing_import_common.go: make dollar sign optional in firstDollarPrice regex
- perplexity_pricing_lib.go: fix column detection to match 'Input ($/1M)' format
- also updated vertex and perplexity baseline snapshots
Previously the test asserted ctyun-live should FAIL, but after
CTYun subscription extension, ctyun-live now passes. Updated
the assertion to match current runtime truth.
- coreshub_pricing_lib.go: CoreHub pricing data extraction and parsing
- import_coreshub_pricing.go: importer with dry_run support
- import_coreshub_pricing_test.go: unit tests for importer
- coreshub_pricing_sample.txt: test fixture
Relax the Tencent catalog plan matcher so monthly promotional plans are parsed by structure instead of a hard-coded plan-name list.
This keeps first-month promotional packages in the catalog and adds a regression sample plus parser test coverage.
Wire the new subscription and official pricing collectors into the daily, real, and intel pipeline entrypoints.
This commit also upgrades Phase 6 verification with recent-window collector classification so gate failures distinguish preconditions from true runtime or provider issues.
Add snapshot, signature, and drift guard support for Vertex AI, Cloudflare Workers AI, and Perplexity API, backed by a queryable audit table and recent-window view.
This commit also wires the audit query layer into daily signal materialization and report generation so structure drift becomes a first-class signal instead of a log-only artifact.
Add plan catalog and subscription schema support, seed baselines, and real importers for core domestic subscriptions plus stable official pricing sources.
This commit also hardens the shared fetch layers so the importers can support live collection and database writes instead of relying on manual placeholders alone.
Tighten real-ingestion success rules, separate scheduled reports from historical rebuilds, and persist source-level runtime audit across daily pipeline runs.
Also add the Phase 5 CI workflow contract plus verification updates and supporting docs so the full uncommitted change set can be validated together.
