@clipwright/mcp-server 0.12.0 → 0.13.0
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/CHANGELOG.md +39 -0
- package/package.json +3 -3
package/CHANGELOG.md
CHANGED
|
@@ -10,6 +10,45 @@ on the client side. Entries marked *Server-side* changed the API itself: every
|
|
|
10
10
|
client version sees them from the day the API was deployed, not from the package
|
|
11
11
|
release they are listed under.
|
|
12
12
|
|
|
13
|
+
## [0.13.0] - 2026-09-13
|
|
14
|
+
|
|
15
|
+
### Changed
|
|
16
|
+
|
|
17
|
+
- `listVoices()`, MCP `list_voices` and `clipwright voices` accept a voice name
|
|
18
|
+
outside the five presets, a `gender` other than `female` or `male`, a voice
|
|
19
|
+
without `language`, and the new optional fields `kind`, `locale`, `accent`,
|
|
20
|
+
`age`, `use_case`, `supported_models` and `verified_models`, plus a top-level
|
|
21
|
+
`models` list. Unknown keys are dropped, as before. Until now the first such
|
|
22
|
+
voice made all three fail. (clipwright#164)
|
|
23
|
+
- A voice name the client does not know goes to the API, which decides.
|
|
24
|
+
`clipwright make --voice` checks only the form of the name: letters, digits,
|
|
25
|
+
`_` and `-`, up to 64 characters. The SDK and MCP `make_ugc` no longer refuse
|
|
26
|
+
such a name before sending it. In `tools/list`, `voice` is a string with a
|
|
27
|
+
pattern instead of an enum of preset names. The API itself is unchanged: a
|
|
28
|
+
name that is not a preset still gets 400 `invalid_request` with the same
|
|
29
|
+
message, before any charge. (clipwright#164)
|
|
30
|
+
- In `VoiceCatalogEntry`, `name` and `gender` are typed `string`, and
|
|
31
|
+
`language` is optional. In `MakeUgcInputArgs`, `voice` is typed `string`.
|
|
32
|
+
(clipwright#164)
|
|
33
|
+
- *Server-side:* quote and run accept a voice from the catalog by its name, as
|
|
34
|
+
well as a preset. A catalog voice speaks `eleven_v3` unless `tts_model` names
|
|
35
|
+
another model, and its script limit follows that model: 5,001 characters
|
|
36
|
+
without `tts_model` get 400 before any charge, as with a preset. A name that
|
|
37
|
+
is neither a preset nor a catalog voice gets 400 `invalid_request` that says
|
|
38
|
+
the voice is unknown; a voice retired from the catalog gets 400 that says it
|
|
39
|
+
was retired. Both point to `list_voices`. The language of a voice limits
|
|
40
|
+
nothing: any catalog voice speaks any language. A run already stored with a
|
|
41
|
+
voice that was later retired still reads, with its warnings.
|
|
42
|
+
(clipwright#164)
|
|
43
|
+
|
|
44
|
+
### Added
|
|
45
|
+
|
|
46
|
+
- `@clipwright/core` exports `VOICE_NAME_PATTERN`, `isVoicePresetName`,
|
|
47
|
+
`serverUgcInputWith` and `voiceModelEntry`. The subpath
|
|
48
|
+
`@clipwright/core/voice-admission` exports `serverUgcInput`, which checks a
|
|
49
|
+
voice name against the catalog, and `catalogVoiceRefusal`. The catalog stays
|
|
50
|
+
out of the main entry point. (clipwright#164)
|
|
51
|
+
|
|
13
52
|
## [0.12.0] - 2026-09-13
|
|
14
53
|
|
|
15
54
|
### Changed
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@clipwright/mcp-server",
|
|
3
|
-
"version": "0.
|
|
3
|
+
"version": "0.13.0",
|
|
4
4
|
"description": "MCP server for Clipwright: generate UGC videos from your coding agent",
|
|
5
5
|
"keywords": [
|
|
6
6
|
"mcp",
|
|
@@ -29,8 +29,8 @@
|
|
|
29
29
|
"dependencies": {
|
|
30
30
|
"@modelcontextprotocol/sdk": "^1.29.0",
|
|
31
31
|
"zod": "^4.4.3",
|
|
32
|
-
"@clipwright/core": "0.
|
|
33
|
-
"@clipwright/sdk": "0.
|
|
32
|
+
"@clipwright/core": "0.13.0",
|
|
33
|
+
"@clipwright/sdk": "0.13.0"
|
|
34
34
|
},
|
|
35
35
|
"devDependencies": {
|
|
36
36
|
"@types/node": "^26.1.1"
|