@lincyaw/skills 1.0.0-rc.60 → 1.0.0-rc.61

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@lincyaw/skills",
3
- "version": "1.0.0-rc.60",
3
+ "version": "1.0.0-rc.61",
4
4
  "description": "Presmium's bundled authoring, animation, presentation and plugin-development skills for coding agents",
5
5
  "license": "Apache-2.0",
6
6
  "repository": {
@@ -36,7 +36,7 @@
36
36
  ],
37
37
  "dependencies": {
38
38
  "jiti": "^2.7.0",
39
- "@lincyaw/kernel": "1.0.0-rc.60"
39
+ "@lincyaw/kernel": "1.0.0-rc.61"
40
40
  },
41
41
  "presmium": {
42
42
  "skills": [
package/skills/index.json CHANGED
@@ -12,8 +12,8 @@
12
12
  "references/publishing.md",
13
13
  "references/writing.md"
14
14
  ],
15
- "version": "1.0.0-rc.60",
16
- "hash": "e7ec95300887502c5e209a6a8307ba70bf8159bf8afb009be1f47fecaf4d50a7"
15
+ "version": "1.0.0-rc.61",
16
+ "hash": "2d6fa793b4b722d80c16ef03afad7a725fcf2a4be5e4f4527b80785e83334770"
17
17
  },
18
18
  {
19
19
  "name": "presmium-animate",
@@ -545,7 +545,7 @@
545
545
  "references/team.md",
546
546
  "references/timeline.md"
547
547
  ],
548
- "version": "1.0.0-rc.60",
548
+ "version": "1.0.0-rc.61",
549
549
  "hash": "76dc5b50143b8134b36c010e5416033652071373500f214f4e7b1792af8a0c4a"
550
550
  },
551
551
  {
@@ -556,7 +556,7 @@
556
556
  "references/plugin.md",
557
557
  "references/testing.md"
558
558
  ],
559
- "version": "1.0.0-rc.60",
559
+ "version": "1.0.0-rc.61",
560
560
  "hash": "af8f48dbe9dbc5bd772fa3d6172ae9c9b96626fec2cd7933b4e08f7c5f154b75"
561
561
  }
562
562
  ]
@@ -41,7 +41,7 @@ Read an existing real plan with `read_production_plan({request:{source}})`; expl
41
41
 
42
42
  For generation, copy the source, IR revision and scope targets from `get_voiceover_status`, then use authorized `start_voiceover` with a client intent. Discover or reconcile through `get_voiceover_jobs/get_voiceover_job`; `cancel_voiceover` can precede admission and remains available when current JSON is invalid. Web and MCP share the same process-local task owner and monotonic sequence. A completed job still needs actual cache/production-plan verification. Audition uses `quote_voice_preview` then separately authorized `preview_voice`; read its bounded audio resource instead of requesting data URIs, and do not describe generation as observed browser playback. The old synchronous `lecture_synthesize` plugin tool is no longer exposed.
43
43
 
44
- For saved audio, read `get_transcription_capabilities` and the actual provider declaration first. A configured engine is not proof that authentication or recognition succeeded. Only `start_transcription` starts paid recognition, after host authorization; never supply `confirmed` or `origin` in model arguments. Use `get_transcriptions` to discover real workspace tasks, `get_transcription` to poll and `get_transcription_result` to read an exact immutable candidate. Read only needed fields with optional `path` (for example `["result", "segments", "0"]`); large values use bounded JSON-text fragments. Continue with `nextOffset` and the same candidate ID, revision and path, then parse the assembled text. No additional source or verification token is needed. Cancellation uses `cancel_transcription`; keep the original intent when an answer is lost. A pre-admission cancellation has no fabricated task, and broken current JSON can still be inspected or cancelled by known task ID.
44
+ For saved audio, read `get_transcription_capabilities` and the actual provider declaration first. A configured engine is not proof that authentication or recognition succeeded. Only `start_transcription` starts paid recognition, after host authorization; never supply `confirmed` or `origin` in model arguments. Use `get_transcriptions` to discover bounded live task summaries; continue with `nextCursor` as `cursor`. New earlier tasks appear on a fresh first page, and a removed cursor returns 409. No source or verification parameter is needed for paging. Use `get_transcription` to poll complete native details and `get_transcription_result` to read an exact immutable candidate. Read only needed fields with optional `path` (for example `["result", "segments", "0"]`); large values use bounded JSON-text fragments. Continue with `nextOffset` and the same candidate ID, revision and path, then parse the assembled text. No additional source or verification token is needed. Cancellation uses `cancel_transcription`; keep the original intent when an answer is lost. A pre-admission cancellation has no fabricated task, and broken current JSON can still be inspected or cancelled by known task ID.
45
45
 
46
46
  Adopt a candidate through `plan_transcription_adoption` and `apply_transcription_adoption`, using the shared source, candidate revision, draft revision and returned plan ID. Explicitly choose stable steps and preserve or resolve existing pronunciation/cue atoms; a conflicted preview has no applicable plan. Do not flatten protected atoms, invent word times or silently change the voice source. Selecting a real TTS voice and generating it remain separate explicit operations. Required agent review remains a host gate, not a payload flag.
47
47