# Changelog of Anime Release & Legal Streaming Discovery Intelligence (`obliging_persimmon_cki/anime-release-legal-streaming-discovery-intelligence`) Actor

- **URL**: https://apify.com/obliging\_persimmon\_cki/anime-release-legal-streaming-discovery-intelligence/changelog.md
- **Full Actor documentation**: https://apify.com/obliging\_persimmon\_cki/anime-release-legal-streaming-discovery-intelligence.md

## Changelog

### 0.1.13 — Final private release candidate

- Prepared the V1 release candidate for manual Store publication without changing the frozen public contract.
- Verified normalized anime resolution, conservative ambiguity handling, release and schedule intelligence, provider-level streaming discovery, optional raw links, normalized relations, and evidence-quality output.
- Verified one `anime-result` PPE event at $0.010 per successfully produced normalized anime result, including bounded spending-limit behavior.
- Completed the final private Cloud regression matrix and documented the release, rollback, source-policy, and post-publication smoke-test procedures.

### 0.1.12 — Phase 5 private release candidate

- Finalized the Store-facing product contract for conservative anime entity resolution, release/schedule intelligence, source relations, provider-level streaming discovery, and evidence-quality reporting.
- Added one `anime-result` PPE event per successfully produced normalized Dataset entity; raw links, provider aggregates, source requests, candidates, ambiguous queries, and zero-result runs are not charged.
- Added bounded spending-limit handling that retains chargeable results and stops before publishing an unpaid result, with `SPENDING_LIMIT_REACHED` in `RUN_SUMMARY`.
- Documented the private release-candidate price examples, AniList operator policy, source limitations, and non-availability semantics.

### Unreleased — Phase 3

- Defined `maxResults` as a cap on distinct high-confidence entities with stable external-ID deduplication.
- Added normalized provider-level streaming aggregates while retaining raw discovery links when requested.
- Removed flattened public aliases and redundant quality/source diagnostics from dataset rows.
- Simplified release, relation, data-quality, input, and Dataset view contracts for Store users.
- Added broad-query, native-title, collision, aggregation, deduplication, and no-streaming fixture coverage.

### Unreleased — Phase 2

- Added normalized entity-match evidence, explicit data-quality components, schedule provenance, and one-level relation normalization.
- Made Jikan conditional fallback-only after AniList cannot produce an unambiguous usable match.
- Added bounded one-retry handling for selected transient HTTP/network failures with `Retry-After` support.
- Added fixture coverage for active/finished schedules, native titles, relation types, malformed streaming evidence, fallback behavior, and retry classification.

### 0.1.0

- Added the Phase 1 discovery-only Actor vertical slice.
- Added conservative title matching and source provenance.
- Added AniList policy gating and replaceable Jikan fallback.
- Added legal streaming discovery links as source evidence without regional verification fields.
- Added Apify Cloud deployment packaging, bounded short-query matching, and production-policy configuration modes.