Releases: scenario-labs/skills
Releases · scenario-labs/skills
Release list
skills: v0.48.0
skills: v0.47.0
0.47.0 (2026-09-25)
Features
- scenario-3d: teach the texture and lighting controls on image-to-3D members (#155) (e12b710)
- scenario-audio: teach two-speaker dialogue, per-member tag syntax, and text caps for speech (#154) (8b6abb9)
- scenario-orbit-views: add new camera angles through a 3D intermediary (#151) (21bccc3)
- scenario-video: teach the whole-song music video lane (#156) (47f3101)
Bug Fixes
skills: v0.46.0
0.46.0 (2026-09-22)
Features
- scenario-game-assets: teach the parallax background lane (#146) (286b77d)
- scenario-model-comparison: add the model comparison skill (#147) (997c715)
- scenario-model-training: teach sample prompts, identity drift, and the LoRA base pairing (#144) (870fe4c)
- skills: teach lyric correction candidates and seeded music takes (#145) (b4c64a2)
Bug Fixes
skills: v0.45.0
0.45.0 (2026-09-21)
Features
- scenario-3d: cover Astra 3D's fit and cost levers, the parts lane, and game-readiness (#141) (75d8a7a)
- scenario-audio: teach the song-extend lane, the stems gap, and per-member seed and batch (#142) (60b6b6d)
- scenario-image: teach landing an exact pixel size, and the blur and literalness fixes (#140) (985465b)
- scenario-patina-retexture: add the PATINA retexture and comparison skill (#137) (7f3e506)
- scenario-video: teach duration floors, frame-anchor rules, reference audio, LoRAs, and retirements (#143) (1d1caf4)
Bug Fixes
skills: v0.44.0
skills: v0.43.0
skills: v0.42.0
0.42.0 (2026-09-16)
Features
- scenario-consistency: add reference-to-video, style-reference, and when-to-train lanes (#114) (9dea74a)
- scenario-moderation: cover video, audio, and IP Detection blocks (#115) (6f8074b)
- scenario-team-admin: add the team administration skill (#118) (8d41021)
- skills: teach shot direction, in-image copy, and frame chaining across image and video skills (#119) (5d223d9)
Bug Fixes
skills: v0.41.1
0.41.1 (2026-09-15)
Bug Fixes
- scenario-sprite-animation: verify extractor frame order instead of assuming time order (#117) (9b56fc1)
- scenario: name the search target, the api_key_create lane, and the full allowlist request (#128) (c82a963)
- scenario: teach per-client re-authentication and the first-call failures usage data surfaced (#116) (0af22c0)
- skills: teach the payload errors usage data surfaces: file field names, image-only inputs, list caps, downloads (#122) (ff10d64)
- skills: verify audio sync after a concat and frame order after an extraction instead of trusting either (#123) (d2a84cf)