@remixmate/cli 0.13.7 → 0.13.9

package/dist/handlers/gen-digital-human.js

@@ -1,39 +1,39 @@
-var k=Object.defineProperty;var f=(t,e)=>k(t,"name",{value:e,configurable:!0});import{mmPost as p,pollUntil as y,resolveHttpContext as U,SkillError as i}from"../http.js";import{getCapabilities as b}from"../capabilities.js";import{emitProgress as m}from"../progress.js";import{isTrue as w,toNumber as M}from"./shared.js";const E=["video","image"],S="hifly";function l(t){return t==null?"":String(t).trim()}f(l,"toId");async function N(t,e){const r=w(e.mine),a=!r&&w(e.all),n=!r&&!a,s={current:1,pageSize:100};n?s.featured=!0:s.source=S,r&&(s.mine=!0),e.gender&&(s.gender=String(e.gender)),e.name&&(s.name=String(e.name));const g=await p(t,"/digital-human/avatar/page",s,{timeoutMs:3e4}),o=g.list??[];if(w(e.json_output)){process.stdout.write(JSON.stringify({avatars:o})+`
-`);return}if(o.length===0){process.stdout.write(n?`⚠️  No curated public avatars match. Try --all for the full inventory, or --mine for your own.
+var U=Object.defineProperty;var w=(t,e)=>U(t,"name",{value:e,configurable:!0});import{mmPost as h,pollUntil as y,resolveHttpContext as b,SkillError as i}from"../http.js";import{getCapabilities as N}from"../capabilities.js";import{emitProgress as v}from"../progress.js";import{isTrue as g,toNumber as I}from"./shared.js";const E=["video","image"],k=["hifly","heygen"],S="hifly";function c(t){return t==null?"":String(t).trim()}w(c,"toId");async function M(t,e){const o=g(e.mine),n=!o&&g(e.all),a=!o&&!n,r={current:1,pageSize:100};a?r.featured=!0:n&&(r.source=c(e.source)||S),o&&(r.mine=!0),e.gender&&(r.gender=String(e.gender)),e.name&&(r.name=String(e.name));const u=await h(t,"/digital-human/avatar/page",r,{timeoutMs:3e4}),l=u.list??[];if(g(e.json_output)){process.stdout.write(JSON.stringify({avatars:l})+`
+`);return}if(l.length===0){process.stdout.write(a?`⚠️  No curated public avatars match. Try --all for the full inventory, or --mine for your own.
 `:`⚠️  No avatars available
-`);return}if(n){process.stdout.write(`
+`);return}if(a){process.stdout.write(`
 ${"ID".padEnd(24)} ${"Name".padEnd(16)} ${"Gender".padEnd(8)} ${"Ratio".padEnd(6)} ${"Scenes".padEnd(22)} Voice
 `),process.stdout.write(`${"-".repeat(24)} ${"-".repeat(16)} ${"-".repeat(8)} ${"-".repeat(6)} ${"-".repeat(22)} ${"-".repeat(12)}
-`);for(const d of o)process.stdout.write(`${l(d.id).padEnd(24)} ${String(d.name??"").padEnd(16)} ${String(d.gender??"").padEnd(8)} ${String(d.aspectRatio??"").padEnd(6)} ${(d.tags??[]).join(",").padEnd(22)} ${d.defaultVoiceId||"-"}
+`);for(const d of l)process.stdout.write(`${c(d.id).padEnd(24)} ${String(d.name??"").padEnd(16)} ${String(d.gender??"").padEnd(8)} ${String(d.aspectRatio??"").padEnd(6)} ${(d.tags??[]).join(",").padEnd(22)} ${d.defaultVoiceId||"-"}
 `)}else{process.stdout.write(`
 ${"ID".padEnd(24)} ${"Name".padEnd(16)} ${"Source".padEnd(14)} ${"Gender".padEnd(8)} ${"Ratio".padEnd(6)} Clone
 `),process.stdout.write(`${"-".repeat(24)} ${"-".repeat(16)} ${"-".repeat(14)} ${"-".repeat(8)} ${"-".repeat(6)} ${"-".repeat(8)}
-`);for(const d of o)process.stdout.write(`${l(d.id).padEnd(24)} ${String(d.name??"").padEnd(16)} ${String(d.source??"").padEnd(14)} ${String(d.gender??"").padEnd(8)} ${String(d.aspectRatio??"").padEnd(6)} ${d.cloneStatus||"-"}
-`)}const u=g.pager?.total??o.length;process.stdout.write(`
-${o.length} avatar(s) shown${u>o.length?` of ${u}; narrow with --name`:""}
-`),process.stdout.write(n?`💡 Use --avatar-id <ID> to pick an avatar; pass its Voice as --voice-id for a matching voice
+`);for(const d of l)process.stdout.write(`${c(d.id).padEnd(24)} ${String(d.name??"").padEnd(16)} ${String(d.source??"").padEnd(14)} ${String(d.gender??"").padEnd(8)} ${String(d.aspectRatio??"").padEnd(6)} ${d.cloneStatus||"-"}
+`)}const s=u.pager?.total??l.length;process.stdout.write(`
+${l.length} avatar(s) shown${s>l.length?` of ${s}; narrow with --name`:""}
+`),process.stdout.write(a?`💡 Use --avatar-id <ID> to pick an avatar; pass its Voice as --voice-id for a matching voice
 `:`💡 Use --avatar-id <ID> to pick an avatar for generation
-`)}f(N,"listAvatars");async function I(t,e){return((await p(t,"/digital-human/avatar/page",{current:1,pageSize:100,mine:!0,source:S},{timeoutMs:3e4})).list??[]).find(n=>l(n.id)===e)??{id:e,source:S}}f(I,"fetchAvatarInfo");function O(t,e){const r=(t.source??"").trim()||S;if(!e.includes(r))throw new i(`❌ avatar uses provider "${r}", which is not available (supported: ${e.join("/")}). Pick another avatar from --list-avatars.`);return r}f(O,"resolveSource");async function T(t,e){const r=l(e.clone_url);if(!/^https?:\/\//i.test(r))throw new i("❌ --clone needs --clone-url <public http(s) URL of a video or portrait image> (local files are not uploaded)");const a=l(e.name);if(!a)throw new i("❌ --clone needs --name <avatar name>");const n=l(e.clone_type)||"video";if(!E.includes(n))throw new i(`❌ --clone-type must be one of ${E.join(" / ")}, received: ${n}`);const s={name:a,sourceType:n,fileUrl:r};e.thumbnail_url&&(s.thumbnailUrl=l(e.thumbnail_url)),e.gender&&(s.gender=l(e.gender)),m({phase:"gen-digital-human:clone-request",sourceType:n});const g=await p(t,"/digital-human/avatar/clone",s,{timeoutMs:6e4}),o=l(g.id);if(!o)throw new i(`❌ could not parse the new avatar id; API returned: ${JSON.stringify(g)}`);m({phase:"gen-digital-human:clone-async",avatarId:o});const u=w(e.json_output);if(e.wait!==void 0&&!w(e.wait)){if(u){process.stdout.write(JSON.stringify({avatarId:o,cloneStatus:g.cloneStatus??"cloning"})+`
+`)}w(M,"listAvatars");async function O(t,e){for(const o of[{mine:!0},{featured:!0}]){const a=((await h(t,"/digital-human/avatar/page",{current:1,pageSize:100,...o},{timeoutMs:3e4})).list??[]).find(r=>c(r.id)===e);if(a)return a}return{id:e,source:S}}w(O,"fetchAvatarInfo");function R(t,e){const o=(t.source??"").trim()||S;if(!e.includes(o))throw new i(`❌ avatar uses provider "${o}", which is not available (supported: ${e.join("/")}). Pick another avatar from --list-avatars.`);return o}w(R,"resolveSource");async function T(t,e){const o=c(e.clone_url);if(!/^https?:\/\//i.test(o))throw new i("❌ --clone needs --clone-url <public http(s) URL of a video or portrait image> (local files are not uploaded)");const n=c(e.name);if(!n)throw new i("❌ --clone needs --name <avatar name>");const a=c(e.clone_type)||"video";if(!E.includes(a))throw new i(`❌ --clone-type must be one of ${E.join(" / ")}, received: ${a}`);const r=c(e.clone_provider)||S;if(!k.includes(r))throw new i(`❌ --clone-provider must be one of ${k.join(" / ")}, received: ${r}`);if(r==="heygen"&&a!=="image")throw new i("❌ HeyGen avatars are created from a photo only; add --clone-type image");const u={name:n,sourceType:a,fileUrl:o};r!==S&&(u.provider=r),e.thumbnail_url&&(u.thumbnailUrl=c(e.thumbnail_url)),e.gender&&(u.gender=c(e.gender)),v({phase:"gen-digital-human:clone-request",sourceType:a,provider:r});const l=await h(t,"/digital-human/avatar/clone",u,{timeoutMs:6e4}),s=c(l.id);if(!s)throw new i(`❌ could not parse the new avatar id; API returned: ${JSON.stringify(l)}`);v({phase:"gen-digital-human:clone-async",avatarId:s});const d=g(e.json_output);if(e.wait!==void 0&&!g(e.wait)){if(d){process.stdout.write(JSON.stringify({avatarId:s,cloneStatus:l.cloneStatus??"cloning"})+`
 `);return}process.stdout.write(`
-🧬 Clone submitted, avatar id: ${o}
-`),process.stdout.write(`💡 Check progress with --clone-status --avatar-id ${o}
-`);return}let d;try{d=await y(async()=>{const c=await p(t,"/digital-human/avatar/clone-status",{id:o},{timeoutMs:3e4}),h=c.cloneStatus??"unknown";if(m({phase:"gen-digital-human:clone-poll",avatarId:o,status:h}),h==="done")return c;if(h==="failed")throw new i(`❌ avatar clone failed: ${c.cloneError||"unknown error"}`);return null},{intervalMs:1e4,timeoutMs:20*6e4})}catch(c){throw c instanceof i&&/polling timed out/.test(c.message)?new i(`❌ avatar ${o} is still cloning; rerun --clone-status --avatar-id ${o} later`):c}if(u){process.stdout.write(JSON.stringify({avatarId:o,cloneStatus:d.cloneStatus})+`
+🧬 Clone submitted, avatar id: ${s}
+`),process.stdout.write(`💡 Check progress with --clone-status --avatar-id ${s}
+`);return}let m;try{m=await y(async()=>{const f=await h(t,"/digital-human/avatar/clone-status",{id:s},{timeoutMs:3e4}),p=f.cloneStatus??"unknown";if(v({phase:"gen-digital-human:clone-poll",avatarId:s,status:p}),p==="done")return f;if(p==="failed")throw new i(`❌ avatar clone failed: ${f.cloneError||"unknown error"}`);return null},{intervalMs:1e4,timeoutMs:20*6e4})}catch(f){throw f instanceof i&&/polling timed out/.test(f.message)?new i(`❌ avatar ${s} is still cloning; rerun --clone-status --avatar-id ${s} later`):f}if(d){process.stdout.write(JSON.stringify({avatarId:s,cloneStatus:m.cloneStatus})+`
 `);return}process.stdout.write(`
-✅ Avatar cloned, id: ${o}
-`),process.stdout.write(`💡 Generate with --avatar-id ${o} --text "..."
-`)}f(T,"cloneAvatar");async function R(t,e,r){const a=await p(t,"/digital-human/avatar/clone-status",{id:e},{timeoutMs:3e4}),n=a.cloneStatus||"unknown";if(r){process.stdout.write(JSON.stringify({avatarId:e,cloneStatus:n,error:a.cloneError||null})+`
+✅ Avatar cloned, id: ${s}
+`),process.stdout.write(`💡 Generate with --avatar-id ${s} --text "..."
+`)}w(T,"cloneAvatar");async function L(t,e,o){const n=await h(t,"/digital-human/avatar/clone-status",{id:e},{timeoutMs:3e4}),a=n.cloneStatus||"unknown";if(o){process.stdout.write(JSON.stringify({avatarId:e,cloneStatus:a,error:n.cloneError||null})+`
 `);return}process.stdout.write(`
-clone status: ${n}
-`),n==="failed"&&process.stdout.write(`   failure reason: ${a.cloneError||"unknown error"}
-`)}f(R,"checkCloneStatus");async function C(t,e,r){const a=await p(t,"/digital-human/video/status",{generationId:e},{timeoutMs:3e4}),n=a.status??"unknown";if(r){process.stdout.write(JSON.stringify({status:n,url:a.videoUrl??null,generationId:e})+`
+clone status: ${a}
+`),a==="failed"&&process.stdout.write(`   failure reason: ${n.cloneError||"unknown error"}
+`)}w(L,"checkCloneStatus");async function j(t,e,o){const n=await h(t,"/digital-human/video/status",{generationId:e},{timeoutMs:3e4}),a=n.status??"unknown";if(o){process.stdout.write(JSON.stringify({status:a,url:n.videoUrl??null,generationId:e})+`
 `);return}if(process.stdout.write(`
-task status: ${n}
-`),a.videoUrl&&process.stdout.write(`   video URL: ${a.videoUrl}
-`),n==="failed"){const s=a.errorMessage??a.error??a.message??"unknown error";process.stdout.write(`   failure reason: ${s}
-`)}}f(C,"checkStatus");const j=30*6e4;async function J(t,e){return await y(async()=>{const a=await p(t,"/digital-human/video/status",{generationId:e},{timeoutMs:3e4}),n=a.status??"unknown";if(m({phase:"gen-digital-human:poll",generationId:e,status:n}),n==="completed")return a.videoUrl??"";if(n==="failed"){const s=a.errorMessage??a.error??a.message??"unknown error";throw new i(`❌ video generation failed: ${s}`)}return null},{intervalMs:5e3,timeoutMs:j}).catch(a=>{throw a instanceof i&&/polling timed out/.test(a.message)?new i(`${a.message}; still generating on the server — resume with --check-status --generation-id ${e}`):a})}f(J,"pollVideoStatus");async function x(t,e){const r=await U(e.skillName,{apiBaseUrl:t.api_base_url,privateToken:t.priv_token,preflightToken:e.auth?.token}),a=w(t.json_output);if(w(t.list_avatars))return N(r,t);if(w(t.clone))return T(r,t);if(w(t.clone_status)){const $=l(t.avatar_id);if(!$)throw new i("❌ clone-status mode requires --avatar-id");return R(r,$,a)}if(w(t.check_status)){const $=M(t.generation_id,"generation_id");if($===void 0)throw new i("❌ check-status mode requires --generation-id");return C(r,$,a)}const n=l(t.avatar_id);if(!n)throw new i("❌ provide --avatar-id to pick the avatar (use --list-avatars to see available avatars)");const s=(t.audio_url??"").trim(),g=t.text??"";if(s&&!/^https?:\/\//i.test(s))throw new i(`❌ --audio-url needs a public http(s) URL (local files are not uploaded), received: ${s}`);if(!s&&!g.trim())throw new i("❌ in TTS mode provide --text, or use --audio-url for audio-driven mode");const o=(await b(t.api_base_url)).digitalHuman;m({phase:"gen-digital-human:resolve-avatar",avatarId:n});const u=await I(r,n);if(u.cloneStatus==="cloning")throw new i(`❌ avatar ${n} is still cloning; check with --clone-status --avatar-id ${n}`);if(u.cloneStatus==="failed")throw new i(`❌ avatar ${n} failed to clone (${u.cloneError||"unknown error"}); clone it again`);const d=O(u,o.sources),c={avatarId:n,source:d};s?c.audioUrl=s:(c.text=g,c.voiceId=t.voice_id??o.defaultVoiceId,t.voice_name&&(c.voiceName=String(t.voice_name))),m({phase:"gen-digital-human:request",avatarId:n,source:d,mode:s?"audio":"tts"});const h=await p(r,"/digital-human/video/generate",c,{timeoutMs:6e4}),v=h.generationId;if(!v)throw new i(`❌ could not parse generationId; API returned: ${JSON.stringify(h)}`);m({phase:"gen-digital-human:async",generationId:v});const _=await J(r,v);if(!_)throw new i(`❌ video generated but no URL was returned; rerun --check-status --generation-id ${v} later`);if(a){process.stdout.write(JSON.stringify({url:_,generationId:v})+`
+task status: ${a}
+`),n.videoUrl&&process.stdout.write(`   video URL: ${n.videoUrl}
+`),a==="failed"){const r=n.errorMessage??n.error??n.message??"unknown error";process.stdout.write(`   failure reason: ${r}
+`)}}w(j,"checkStatus");const C=30*6e4;async function D(t,e){return await y(async()=>{const n=await h(t,"/digital-human/video/status",{generationId:e},{timeoutMs:3e4}),a=n.status??"unknown";if(v({phase:"gen-digital-human:poll",generationId:e,status:a}),a==="completed")return n.videoUrl??"";if(a==="failed"){const r=n.errorMessage??n.error??n.message??"unknown error";throw new i(`❌ video generation failed: ${r}`)}return null},{intervalMs:5e3,timeoutMs:C}).catch(n=>{throw n instanceof i&&/polling timed out/.test(n.message)?new i(`${n.message}; still generating on the server — resume with --check-status --generation-id ${e}`):n})}w(D,"pollVideoStatus");async function q(t,e){const o=await b(e.skillName,{apiBaseUrl:t.api_base_url,privateToken:t.priv_token,preflightToken:e.auth?.token}),n=g(t.json_output);if(g(t.list_avatars))return M(o,t);if(g(t.clone))return T(o,t);if(g(t.clone_status)){const $=c(t.avatar_id);if(!$)throw new i("❌ clone-status mode requires --avatar-id");return L(o,$,n)}if(g(t.check_status)){const $=I(t.generation_id,"generation_id");if($===void 0)throw new i("❌ check-status mode requires --generation-id");return j(o,$,n)}const a=c(t.avatar_id);if(!a)throw new i("❌ provide --avatar-id to pick the avatar (use --list-avatars to see available avatars)");const r=(t.audio_url??"").trim(),u=t.text??"";if(r&&!/^https?:\/\//i.test(r))throw new i(`❌ --audio-url needs a public http(s) URL (local files are not uploaded), received: ${r}`);if(!r&&!u.trim())throw new i("❌ in TTS mode provide --text, or use --audio-url for audio-driven mode");const l=(await N(t.api_base_url)).digitalHuman;v({phase:"gen-digital-human:resolve-avatar",avatarId:a});const s=await O(o,a);if(s.cloneStatus==="cloning")throw new i(`❌ avatar ${a} is still cloning; check with --clone-status --avatar-id ${a}`);if(s.cloneStatus==="failed")throw new i(`❌ avatar ${a} failed to clone (${s.cloneError||"unknown error"}); clone it again`);const d=R(s,l.sources),m={avatarId:a,source:d};r?m.audioUrl=r:(m.text=u,m.voiceId=c(t.voice_id)||c(s.defaultVoiceId)||l.defaultVoiceId,t.voice_name&&(m.voiceName=String(t.voice_name))),v({phase:"gen-digital-human:request",avatarId:a,source:d,mode:r?"audio":"tts"});const f=await h(o,"/digital-human/video/generate",m,{timeoutMs:6e4}),p=f.generationId;if(!p)throw new i(`❌ could not parse generationId; API returned: ${JSON.stringify(f)}`);v({phase:"gen-digital-human:async",generationId:p});const _=await D(o,p);if(!_)throw new i(`❌ video generated but no URL was returned; rerun --check-status --generation-id ${p} later`);if(n){process.stdout.write(JSON.stringify({url:_,generationId:p})+`
 `);return}process.stdout.write(`
 🔗 Video URL (renders directly, no download needed):
 ${_}
 `),process.stdout.write(`
 🎉 Digital-human video generation done!
-`)}f(x,"genDigitalHuman");export{x as genDigitalHuman};
+`)}w(q,"genDigitalHuman");export{q as genDigitalHuman};

package/dist/handlers/gen-voice.js

@@ -1,19 +1,23 @@
-var w=Object.defineProperty;var d=(e,i)=>w(e,"name",{value:i,configurable:!0});import{mmPost as m,resolveHttpContext as v,SkillError as c}from"../http.js";import{getCapabilities as y}from"../capabilities.js";import{emitProgress as x}from"../progress.js";import{isTrue as u,toNumber as _}from"./shared.js";const b=[{id:"Chinese (Mandarin)_Male_Announcer",language:"zh",name:"Mandarin Male Announcer (legacy default)"},{id:"english-male-friendly-01",language:"en",name:"Friendly Male (English, TBD)"}];async function $(e){const o=(await m(e,"/voice/page",{current:1,pageSize:100},{timeoutMs:3e4})).list??[];if(o.length===0){process.stdout.write(`⚠️  No voices available
-`);return}process.stdout.write(`
-${"Voice ID".padEnd(30)} Name
-`),process.stdout.write(`${"-".repeat(30)} ${"-".repeat(20)}
-`);for(const s of o)process.stdout.write(`${(s.uniqId??"").padEnd(30)} ${s.name??""}
-`);process.stdout.write(`
-${o.length} voice(s) available
-`),process.stdout.write(`💡 Use --voice-id <id> to select a voice for synthesis
-`)}d($,"listVoicesRemote");function k(){process.stderr.write(`⚠️  --local prints a hardcoded 2-entry fallback list, NOT the available voices. Drop --local to query the live catalog.
-`);for(const e of b)process.stdout.write(`${e.id}	${e.language}	${e.name}
-`)}d(k,"listVoicesLocal");async function L(e,i){if(u(e.list_voices)){if(u(e.local))return k();const t=await v(i.skillName,{apiBaseUrl:e.api_base_url,privateToken:e.priv_token,preflightToken:i.auth?.token});return $(t)}const o=e.text;if(!o||typeof o!="string"||!o.trim())throw new c("❌ Please provide --text to specify the text to synthesize");const s=e.api_base_url,n=(await y(s)).voice,p=e.voice_id??n.defaultVoiceId,r=_(e.speed,"speed")??1;if(r<n.speedMin||r>n.speedMax)throw new c(`❌ speed must be in the range ${n.speedMin}~${n.speedMax}`);const g=e.language_boost?.trim()||void 0,h=await v(i.skillName,{apiBaseUrl:s,privateToken:e.priv_token,preflightToken:i.auth?.token});x({phase:"gen-voice:request",voiceId:p,speed:r});const a=await m(h,"/tool/tts",{provider:n.provider,text:o,voiceSetting:{voiceId:p,speed:r},outputFormat:"url",...g?{languageBoost:g}:{}},{timeoutMs:12e4}),l=a.audio;if(!l)throw new c(`❌ API response did not contain an audio URL: ${JSON.stringify(a)}`);if(u(e.json_output)){process.stdout.write(JSON.stringify({url:l,audio_length_ms:a.extraInfo?.audioLength??null,subtitles:(a.subtitles??[]).map(t=>({text:t.text??"",timeBegin:t.timeBegin??0,timeEnd:t.timeEnd??0,...typeof t.textBegin=="number"&&typeof t.textEnd=="number"?{textBegin:t.textBegin,textEnd:t.textEnd}:{}}))})+`
-`);return}const f=a.extraInfo?.audioLength;typeof f=="number"&&process.stdout.write(`
-🕐 Audio length: ${(f/1e3).toFixed(1)} s
+var T=Object.defineProperty;var l=(e,o)=>T(e,"name",{value:o,configurable:!0});import{mmPost as w,resolveHttpContext as $,SkillError as h}from"../http.js";import{getCapabilities as b}from"../capabilities.js";import{emitProgress as A}from"../progress.js";import{isTrue as m,toNumber as y}from"./shared.js";const L=[{id:"Chinese (Mandarin)_Male_Announcer",language:"zh",name:"Mandarin Male Announcer (legacy default)"},{id:"english-male-friendly-01",language:"en",name:"Friendly Male (English, TBD)"}],E=100,_=1,x=3,M={male:"男",m:"男",man:"男",female:"女",f:"女",woman:"女"};function f(e){if(e==null)return;const o=String(e).trim();return o||void 0}l(f,"optString");function v(e,o){return(e.tagItems??[]).filter(t=>t.category===o&&t.name).map(t=>t.name).join("/")}l(v,"tagsOf");function N(e){const o=y(e.page,"page")??1;if(!Number.isInteger(o)||o<1)throw new h(`❌ page must be a positive integer, received: ${e.page}`);const t={current:o,pageSize:E},p=f(e.name);p&&(t.name=p);const r=f(e.language);r&&(t.language=r);const i=f(e.gender);if(i&&(t.gender=M[i.toLowerCase()]??i),m(e.mine))return{path:"/voice/collect/myPage",body:t};const s=f(e.provider)?.toLowerCase();s&&(t.provider=s);const a=f(e.voice_type)?.toLowerCase();return a&&(t.voiceType=a),{path:"/voice/page",body:t}}l(N,"buildVoiceListRequest");async function k(e,o){const{path:t,body:p}=N(o),r=await w(e,t,p,{timeoutMs:3e4}),i=r.list??[],s=t!=="/voice/page",a=Number(p.current),g=r.pager?.total??i.length;if(m(o.json_output)){process.stdout.write(JSON.stringify({page:a,total:g,voices:i.map(n=>({id:n.uniqId??"",name:n.name??"",provider:n.provider??"",type:n.voiceType??"",gender:v(n,x),language:v(n,_)}))})+`
+`);return}if(i.length===0){process.stdout.write(a>1?`⚠️  No voices on page ${a} (total ${g})
+`:s?`⚠️  You have no cloned voices
+`:`⚠️  No voices match these filters
+`);return}const d=Math.min(48,Math.max(30,...i.map(n=>(n.uniqId??"").length)));process.stdout.write(`
+${"Voice ID".padEnd(d)} ${"Provider".padEnd(10)} ${"Type".padEnd(6)} ${"Gender".padEnd(6)} ${"Language".padEnd(12)} Name
+`),process.stdout.write(`${"-".repeat(d)} ${"-".repeat(10)} ${"-".repeat(6)} ${"-".repeat(6)} ${"-".repeat(12)} ${"-".repeat(20)}
+`);for(const n of i)process.stdout.write(`${(n.uniqId??"").padEnd(d)} ${(n.provider??"").padEnd(10)} ${(n.voiceType??"").padEnd(6)} ${v(n,x).padEnd(6)} ${v(n,_).padEnd(12)} ${n.name??""}
+`);const u=(a-1)*E+i.length;process.stdout.write(`
+${i.length} voice(s) on page ${a}, ${g} in total
+`),g>u&&process.stdout.write(`💡 More voices: add --page ${a+1}, or narrow with --provider / --gender / --language / --name
+`),process.stdout.write(`💡 Use --voice-id <id> to select a voice; the provider is picked from the voice automatically
+`)}l(k,"listVoicesRemote");function C(){process.stderr.write(`⚠️  --local prints a hardcoded 2-entry fallback list, NOT the available voices. Drop --local to query the live catalog.
+`);for(const e of L)process.stdout.write(`${e.id}	${e.language}	${e.name}
+`)}l(C,"listVoicesLocal");async function R(e,o){if(m(e.list_voices)){if(m(e.local))return C();const c=await $(o.skillName,{apiBaseUrl:e.api_base_url,privateToken:e.priv_token,preflightToken:o.auth?.token});return k(c,e)}const t=e.text;if(!t||typeof t!="string"||!t.trim())throw new h("❌ Please provide --text to specify the text to synthesize");const p=e.api_base_url,r=(await b(p)).voice,i=e.voice_id??r.defaultVoiceId,s=y(e.speed,"speed")??1;if(s<r.speedMin||s>r.speedMax)throw new h(`❌ speed must be in the range ${r.speedMin}~${r.speedMax}`);const a=e.language_boost?.trim()||void 0,g=await $(o.skillName,{apiBaseUrl:p,privateToken:e.priv_token,preflightToken:o.auth?.token});A({phase:"gen-voice:request",voiceId:i,speed:s});const d=await w(g,"/tool/tts",{provider:r.provider,text:t,voiceSetting:{voiceId:i,speed:s},outputFormat:"url",...a?{languageBoost:a}:{}},{timeoutMs:12e4}),u=d.audio;if(!u)throw new h(`❌ API response did not contain an audio URL: ${JSON.stringify(d)}`);if(m(e.json_output)){process.stdout.write(JSON.stringify({url:u,audio_length_ms:d.extraInfo?.audioLength??null,subtitles:(d.subtitles??[]).map(c=>({text:c.text??"",timeBegin:c.timeBegin??0,timeEnd:c.timeEnd??0,...typeof c.textBegin=="number"&&typeof c.textEnd=="number"?{textBegin:c.textBegin,textEnd:c.textEnd}:{}}))})+`
+`);return}const n=d.extraInfo?.audioLength;typeof n=="number"&&process.stdout.write(`
+🕐 Audio length: ${(n/1e3).toFixed(1)} s
 `),process.stdout.write(`
 🔗 Audio URL (plays directly, no download needed):
-${l}
+${u}
 `),process.stdout.write(`
 🎉 TTS synthesis done!
-`)}d(L,"genVoice");export{L as genVoice};
+`)}l(R,"genVoice");export{N as buildVoiceListRequest,R as genVoice};

package/dist/manifest.json

@@ -1,7 +1,7 @@
 {
   "schemaVersion": 1,
-  "version": "0.13.7",
-  "generatedAt": "2026-10-30T09:41:49.930Z",
+  "version": "0.13.9",
+  "generatedAt": "2026-11-03T15:23:12.126Z",
   "skills": [
     {
       "id": "doc-parse",
@@ -203,7 +203,15 @@
           },
           "all": {
             "type": "boolean",
-            "description": "With list_avatars=true: list the full public HiFly inventory (500+, mostly uncurated, no scene tags) instead of the curated public avatars. Only for when the curated set has nothing suitable."
+            "description": "With list_avatars=true: list one provider's full public inventory (HiFly by default, 500+, mostly uncurated, no scene tags) instead of the curated public avatars. Only for when the curated set has nothing suitable."
+          },
+          "source": {
+            "type": "string",
+            "enum": [
+              "hifly",
+              "heygen"
+            ],
+            "description": "With list_avatars=true and all=true: which provider's full public inventory to list. Defaults to hifly. heygen is only available when the capabilities catalog lists it."
           },
           "name": {
             "type": "string",
@@ -223,7 +231,7 @@
           },
           "clone": {
             "type": "boolean",
-            "description": "Clone mode: create a custom HiFly avatar from a video (free) or a portrait image (costs credits per clone, refunded automatically if cloning fails; see the pricing page). Requires clone_url and name. Waits until cloning finishes unless wait=false."
+            "description": "Clone mode: create a custom avatar (HiFly by default, or HeyGen with clone_provider=heygen) from a video (free) or a portrait image (costs credits per clone, refunded automatically if cloning fails; see the pricing page). Requires clone_url and name. Waits until cloning finishes unless wait=false."
           },
           "clone_url": {
             "type": "string",
@@ -237,6 +245,14 @@
             ],
             "description": "With clone=true: source type. Defaults to video."
           },
+          "clone_provider": {
+            "type": "string",
+            "enum": [
+              "hifly",
+              "heygen"
+            ],
+            "description": "With clone=true: where to create the avatar. hifly (default) accepts a video or a photo; heygen accepts a photo only (clone_type=image, jpg/png ≤ 32MB) and gives more natural lip-sync for non-Chinese speech at a higher per-second price. Only use heygen when the capabilities catalog lists it."
+          },
           "thumbnail_url": {
             "type": "string",
             "description": "With clone=true: optional cover image URL shown in avatar lists (image clones default to the image itself). The output aspect ratio always follows the cloned source; it cannot be chosen at generation time."
@@ -259,7 +275,7 @@
           },
           "voice_id": {
             "type": "string",
-            "description": "Voice id (TTS mode). Shares the catalog with gen-voice (including cloned voices) — call gen_voice with list_voices=true to see available ids rather than inventing one."
+            "description": "Voice id (TTS mode). Optional: defaults to the avatar's own voice (a curated avatar's suggested voice, or a HeyGen digital twin's cloned voice), else the service default. Shares the catalog with gen-voice (any provider, including cloned voices) — call gen_voice with list_voices=true to see available ids rather than inventing one."
           },
           "voice_name": {
             "type": "string",
@@ -294,11 +310,13 @@
           "clone",
           "clone_url",
           "clone_type",
+          "clone_provider",
           "thumbnail_url",
           "wait",
           "clone_status",
           "mine",
           "all",
+          "source",
           "name",
           "gender",
           "check_status",
@@ -614,7 +632,7 @@
               "on",
               "off"
             ],
-            "description": "Narration on / off, overriding the template's needsNarration (auto = the template decides). Templates where narration is optional (image sequences, galton-board) default to none — **must** pass `on` when the replicate card says `旁白: <名称>（narrated）`, otherwise the narration you write is dropped without an error and the video comes back silent. `off` drops narration from a narrated template."
+            "description": "Narration on / off, overriding the template's needsNarration (auto = the template decides). Templates declaring narration optional default to none — **must** pass `on` when the replicate card says `旁白: <名称>（narrated）`, otherwise the narration you write is dropped without an error and the video comes back silent. `off` drops narration from a narrated template."
           },
           "background_asset": {
             "type": "string",
@@ -925,11 +943,11 @@
       "toolName": "gen_voice",
       "tier": "atomic",
       "category": "asset",
-      "title": "Text-to-Speech (Minimax)",
-      "summary": "Text-to-speech (TTS): synthesize narration audio from text via the Minimax TTS model. Returns the persisted audio URL — no download needed.",
+      "title": "Text-to-Speech",
+      "summary": "Text-to-speech (TTS): synthesize narration audio from text. The voice catalog spans several providers (MiniMax, 豆包/Volcengine, HeyGen, and the caller's own cloned voices); the provider is picked from the voice id automatically. Returns the persisted audio URL — no download needed.",
       "triggers": [
         "AI voice-over, TTS, text-to-speech, generate narration audio",
-        "Use Minimax for speech synthesis",
+        "Use MiniMax / 豆包 (Doubao) / HeyGen voices for speech synthesis",
         "Read a piece of text aloud / produce an audio file from text"
       ],
       "entry": {
@@ -951,15 +969,57 @@
           },
           "voice_id": {
             "type": "string",
-            "description": "Voice id. Default 'Chinese (Mandarin)_Male_Announcer'. When unsure, call with list_voices=true first to see what's available — do not invent ids, and do not declare a voice unavailable unless it is missing from that list."
+            "description": "Voice id from list_voices (any provider; synthesis routes to the voice's own provider). Default 'Chinese (Mandarin)_Male_Announcer'. When unsure, call with list_voices=true first — do not invent ids, and do not declare a voice unavailable unless it is missing from the filtered list."
           },
           "speed": {
             "type": "number",
-            "description": "Speech rate, 0.5-2.0"
+            "description": "Speech rate, 0.5-2.0. Some HeyGen cloned voices ignore it."
           },
           "list_voices": {
             "type": "boolean",
-            "description": "List the available voices and exit. This queries the live voice service and is the authoritative catalog — one call is enough, and its absence from this list is the only evidence that a voice id is invalid."
+            "description": "List voices and exit (one page of up to 100, with provider / type / gender / language columns and the total). This queries the live voice service and is the authoritative catalog. The public catalog has a few hundred voices, so narrow it with provider / gender / language / name, or page through with page; mine=true lists the caller's own cloned voices instead."
+          },
+          "mine": {
+            "type": "boolean",
+            "description": "With list_voices=true: list the caller's own cloned voices instead of the public catalog (provider / voice_type do not apply)."
+          },
+          "provider": {
+            "type": "string",
+            "enum": [
+              "minimax",
+              "volcengine",
+              "heygen"
+            ],
+            "description": "With list_voices=true: only voices from this provider (volcengine = 豆包). Not needed for synthesis."
+          },
+          "voice_type": {
+            "type": "string",
+            "enum": [
+              "system",
+              "clone"
+            ],
+            "description": "With list_voices=true: system (stock) or clone voices."
+          },
+          "gender": {
+            "type": "string",
+            "enum": [
+              "male",
+              "female"
+            ],
+            "description": "With list_voices=true: filter by gender."
+          },
+          "language": {
+            "type": "string",
+            "description": "With list_voices=true: filter by language tag, exact tag as shown in the Language column (e.g. 中文-普通话, 中文-粤语, 英语, 日语)."
+          },
+          "name": {
+            "type": "string",
+            "description": "With list_voices=true: fuzzy-filter by voice name."
+          },
+          "page": {
+            "type": "integer",
+            "minimum": 1,
+            "description": "With list_voices=true: page number (100 per page). Default 1."
           },
           "json_output": {
             "type": "boolean",
@@ -978,7 +1038,14 @@
         ],
         "hidden": [
           "json_output",
-          "list_voices"
+          "list_voices",
+          "mine",
+          "provider",
+          "voice_type",
+          "gender",
+          "language",
+          "name",
+          "page"
         ]
       }
     },

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@remixmate/cli",
-  "version": "0.13.7",
+  "version": "0.13.9",
   "description": "AI media generation skills for Claude Code / Codex — 12 skills covering image, video, voice, digital human, web screenshot, web recording, script, template registry, rendering, Jianying export, and video deconstruction.",
   "type": "module",
   "main": "dist/index.js",

package/skills/gen-digital-human/SKILL.md

@@ -19,7 +19,7 @@ triggers:
 
 # Digital-Human Skill
 
-Wraps ab-api's digital-human endpoints (the same ones the web studio uses), authenticated with the **Tianyan privateToken**. The provider is **HiFly**: avatars are cloned from a video or photo, then driven by audio. Generation is async — the handler submits the task and polls `/digital-human/video/status` until completion.
+Wraps ab-api's digital-human endpoints (the same ones the web studio uses), authenticated with the **Tianyan privateToken**. Providers are **HiFly** (default) and, when the capabilities catalog lists it, **HeyGen**: avatars are cloned from a video or photo (HeyGen: photo only), then driven by audio. Each avatar belongs to one provider and always renders there — you never pick the provider at generation time. HeyGen costs more per second (the curated list and the studio picker mark HeyGen avatars); prefer it only for non-Chinese speech or when the user asks. Generation is async — the handler submits the task and polls `/digital-human/video/status` until completion.
 
 ## Capabilities
 
@@ -57,6 +57,7 @@ remixmate gen-digital-human --list-avatars                    # curated public a
 remixmate gen-digital-human --list-avatars --gender female    # curated, filtered by gender
 remixmate gen-digital-human --list-avatars --mine             # your cloned avatars
 remixmate gen-digital-human --list-avatars --all --name "教师" # full HiFly inventory (500+; always narrow with --name)
+remixmate gen-digital-human --list-avatars --all --source heygen --name "Anna" # full HeyGen public inventory
 ```
 
 The default list is the **curated public library**: avatars an admin picked, with a cover, scene tags (`education` / `medical` / `news` / `business` / `culture` / `outdoor` / `lifestyle` / `festival` / `ecommerce`) and usually a suggested voice. Prefer it — pick the avatar whose scenes match the script, and pass its `Voice` column as `--voice-id` unless the user chose a voice. Fall back to `--all` only when nothing curated fits; those entries have no tags or suggested voice.
@@ -94,6 +95,9 @@ remixmate gen-digital-human --clone --name "My presenter" --clone-url "https://e
 # Portrait photo clone (charged per clone)
 remixmate gen-digital-human --clone --clone-type image --name "My presenter" --clone-url "https://example.com/me.jpg"
 
+# HeyGen photo avatar (photo only, jpg/png ≤ 32MB; more natural lip-sync for English and other languages)
+remixmate gen-digital-human --clone --clone-type image --clone-provider heygen --name "My presenter" --clone-url "https://example.com/me.jpg"
+
 # Submit without waiting, then poll
 remixmate gen-digital-human --clone --wait false --name "My presenter" --clone-url "https://example.com/me.mp4"
 remixmate gen-digital-human --clone-status --avatar-id dh_xxx
@@ -109,12 +113,14 @@ remixmate gen-digital-human --clone-status --avatar-id dh_xxx
 | `--name` | list / clone | Fuzzy filter; with `--clone` the new avatar's name (required) | — |
 | `--avatar-id` | generate / clone-status | Avatar id (required) | — |
 | `-t` / `--text` | generate | Script (text-driven) | — |
-| `--voice-id` | generate | Voice id (text-driven) | `male-qn-qingse` |
+| `--voice-id` | generate | Voice id (text-driven) | the avatar's own voice (curated suggestion / HeyGen twin's cloned voice), else the service default |
 | `--voice-name` | generate | Voice display name (recordkeeping only) | none |
 | `--audio-url` | generate | Audio URL (audio-driven) | none |
 | `--check-status` / `--generation-id` | status | Poll a generation job | — |
 | `--clone` / `--clone-url` | clone | Clone mode and source URL (required) | — |
 | `--clone-type` | clone | `video` / `image` | `video` |
+| `--clone-provider` | clone | `hifly` / `heygen` (heygen needs `--clone-type image`) | `hifly` |
+| `--source` | list with `--all` | Which provider's full inventory: `hifly` / `heygen` | `hifly` |
 | `--thumbnail-url` | clone | Cover shown in lists | image itself / none |
 | `--wait` | clone | `false` to return right after submission | wait |
 | `--clone-status` | clone-status | Report clone status of `--avatar-id` | — |

package/skills/gen-digital-human/skill.json

@@ -29,7 +29,15 @@
       },
       "all": {
         "type": "boolean",
-        "description": "With list_avatars=true: list the full public HiFly inventory (500+, mostly uncurated, no scene tags) instead of the curated public avatars. Only for when the curated set has nothing suitable."
+        "description": "With list_avatars=true: list one provider's full public inventory (HiFly by default, 500+, mostly uncurated, no scene tags) instead of the curated public avatars. Only for when the curated set has nothing suitable."
+      },
+      "source": {
+        "type": "string",
+        "enum": [
+          "hifly",
+          "heygen"
+        ],
+        "description": "With list_avatars=true and all=true: which provider's full public inventory to list. Defaults to hifly. heygen is only available when the capabilities catalog lists it."
       },
       "name": {
         "type": "string",
@@ -49,7 +57,7 @@
       },
       "clone": {
         "type": "boolean",
-        "description": "Clone mode: create a custom HiFly avatar from a video (free) or a portrait image (costs credits per clone, refunded automatically if cloning fails; see the pricing page). Requires clone_url and name. Waits until cloning finishes unless wait=false."
+        "description": "Clone mode: create a custom avatar (HiFly by default, or HeyGen with clone_provider=heygen) from a video (free) or a portrait image (costs credits per clone, refunded automatically if cloning fails; see the pricing page). Requires clone_url and name. Waits until cloning finishes unless wait=false."
       },
       "clone_url": {
         "type": "string",
@@ -63,6 +71,14 @@
         ],
         "description": "With clone=true: source type. Defaults to video."
       },
+      "clone_provider": {
+        "type": "string",
+        "enum": [
+          "hifly",
+          "heygen"
+        ],
+        "description": "With clone=true: where to create the avatar. hifly (default) accepts a video or a photo; heygen accepts a photo only (clone_type=image, jpg/png ≤ 32MB) and gives more natural lip-sync for non-Chinese speech at a higher per-second price. Only use heygen when the capabilities catalog lists it."
+      },
       "thumbnail_url": {
         "type": "string",
         "description": "With clone=true: optional cover image URL shown in avatar lists (image clones default to the image itself). The output aspect ratio always follows the cloned source; it cannot be chosen at generation time."
@@ -85,7 +101,7 @@
       },
       "voice_id": {
         "type": "string",
-        "description": "Voice id (TTS mode). Shares the catalog with gen-voice (including cloned voices) — call gen_voice with list_voices=true to see available ids rather than inventing one."
+        "description": "Voice id (TTS mode). Optional: defaults to the avatar's own voice (a curated avatar's suggested voice, or a HeyGen digital twin's cloned voice), else the service default. Shares the catalog with gen-voice (any provider, including cloned voices) — call gen_voice with list_voices=true to see available ids rather than inventing one."
       },
       "voice_name": {
         "type": "string",
@@ -120,11 +136,13 @@
       "clone",
       "clone_url",
       "clone_type",
+      "clone_provider",
       "thumbnail_url",
       "wait",
       "clone_status",
       "mine",
       "all",
+      "source",
       "name",
       "gender",
       "check_status",

package/skills/gen-script/SKILL.md

@@ -491,7 +491,7 @@ python3 <SkillDir>/scripts/gen_script.py \
 | `--payload KEY=VALUE` | Template-specific card choice, repeatable. `KEY` is a dotted path in `customPayloadSchema` (or `capabilities.frameLevelKeys`), e.g. `<field>=<value>`, `<object>.<field>=<value>`. Value is converted to the declared type (true/false, numbers) and checked against enum / range; undeclared keys are rejected. Written to every scene with sibling keys kept, applied before fit-caption duration estimation. | template default |
 | `--ending-payload KEY=VALUE` | Same as `--payload`, but written only to the ending scene, e.g. `<object>.<field>=<closing line>`. On countdown templates it also appends an ending scene after #1 (it carries the #1 index; fill its narration and the item field `capabilities.countdown.itemProp` names). Errors when the script has no ending scene. | none |
 | `--avatar-id` | Digital-human avatar id (`dh_xxx`) for templates declaring `capabilities.digitalHuman`. Adds asset `asset-avatar` (`type: avatar`, `source: gen-digital-human`, `payload.avatarId`) and sets `visuals.avatar.assetRef` on every scene whose slot reads it — one continuous presenter driven by the narration. Placement: template default, or `--payload digitalHuman.<field>=…` (a photo-only presenter is `--payload digitalHuman.posterUrl=<url>` without this flag). | none |
-| `--motion` | Camera motion for every scene (`animationHints.motion`): `static` / `kenburns-in` / `kenburns-out` / `pan-left` / `pan-right` / `float`. `auto` removes the motion from every scene so the template picks (image-slide rotates push / pan / pull / pan); nothing is locked then. | skeleton alternates in / out |
+| `--motion` | Camera motion for every scene (`animationHints.motion`): `static` / `kenburns-in` / `kenburns-out` / `pan-left` / `pan-right` / `float`. `auto` removes the motion from every scene so the template picks its own motion; nothing is locked then. | skeleton alternates in / out |
 | `--font-id` | Font library key (`font.uniq_id`), written to `global.font.fontId`; ab-render resolves and installs the real font at render time. Pass it whenever the user picked a font — the replicate card shows it as `名字（id）`, e.g. `字体: <名字>（<uniqId>）`; pass the id in the parentheses. Never pass a CSS family or the display name: an unknown key falls back to the template font with no error. | template font |
 | `--font-name` | Display name of the picked font → `global.font.fontName` (display only, never used for rendering). Pass it with `--font-id`. | — |
 | `--list-fonts` | Print the live font library (`/font/page`: id, name, lang, category, license) and exit; no script is generated and `--topic` is not needed. The ids it prints are the only valid `--font-id` values — use it when the user asks which fonts exist or names a font without its id. | — |
@@ -510,7 +510,7 @@ python3 <SkillDir>/scripts/gen_script.py \
 
 Values written by `--background-preset`, `--background-asset`, `--payload`, `--ending-payload`, `--avatar-id` and `--motion` are listed in `meta.lockedScenePaths` (scene-relative paths such as `customPayload.<object>.<field>`, `animationHints.motion`). When the agent later saves its own `dsl_json` in merge mode, ab-agent restores those paths from this skeleton after merging, so a model-written `customPayload` object cannot wipe a card choice. Do not hand-write these keys in `dsl_json`; pass the flag. Templates that author their own scenes (`meta.authoredScenes`) are not merged: there ab-agent writes each locked value that the skeleton holds on every scene (with one value) onto every submitted scene, and adds the skeleton assets those values reference (`asset-background`, `asset-avatar`) when the submitted `assets` lacks them.
 
-For image-to-video replicate cards, map **转场效果 / Transition effect** to both `--payload sequence.defaultEffect=<id>` and `--payload sequence.overrideItemEffects=true`. Use the id in parentheses (`fade`, `wipe-left`, `slide-left`, `none`), preserving the chosen preset and all typography. **跟随样片 / Follow example (`follow`)** omits both overrides; `follow` is not a valid effect. Map **转场时长 / Transition duration** to `--payload sequence.transitionMs=<milliseconds>` (e.g. `0.6 秒（600）` → `600`); omit when following the sample or using `none`. These payload keys are locked into every generated scene, so do not re-author them in a saved DSL.
+For replicate cards exposing transition effects or duration, read the template's `customPayloadSchema` and `llmHint` for the exact field paths, units, allowed values and any paired override switch. Pass each explicit selection through `--payload <declared-path>=<value>`, preserving the chosen preset and typography; following the sample omits the overrides. These payload keys are locked into every generated scene, so do not re-author them in a saved DSL.
 
 ## DSL generation principles
 

package/skills/gen-script/scripts/gen_script.py

@@ -643,6 +643,8 @@ def _template_needs_image(template_config: dict | None) -> bool:
 
 
 
+
+
 
 
 
@@ -707,6 +709,7 @@ def _resolve_contract(template_config: dict | None) -> dict:
         "duration_strategy": duration_strategy,
         "duration_model": caps.get("durationModel"),
         "payload_defaults": payload_defaults,
+        "media_binding": caps.get("mediaBinding"),
         "primary_visual_type": primary_visual_type,
         "background_strategy": "shared" if caps.get("backgroundStrategy") == "shared" else "per-scene",
         "default_layout": caps.get("defaultLayout"),
@@ -1275,6 +1278,8 @@ def _media_option_choice(option_schema: dict, value: str | None, flag: str,
 
 
 
+
+
 
 _MEDIA_OPTION_FLAGS = {"preset": "--sequence-preset", "fit": "--sequence-fit"}
 _LEGACY_MEDIA_BINDINGS = {
@@ -2493,7 +2498,7 @@ def _attach_narration(scene: dict, narration_text: str, narration_asset_id: str
 def _visual_scene(
         ctx: _SkeletonContext, idx: int, scene_id: str, plan: dict, visual_asset_id: str,
 ) -> dict:
-    return {
+    scene = {
         "id": scene_id,
         "purpose": plan["purpose"],
         "duration": plan["duration"],
@@ -2507,6 +2512,24 @@ def _visual_scene(
     }
 
 
+
+    binding = ctx.contract.get("media_binding")
+    if isinstance(binding, dict) and "assetRefKey" in binding:
+        path = binding.get("itemsPath")
+        ref_key = binding.get("assetRefKey")
+        if not isinstance(path, str) or not all(path.split(".")) or not isinstance(ref_key, str) or not ref_key:
+            raise ValueError("mediaBinding needs non-empty itemsPath and assetRefKey")
+        item = deepcopy(binding.get("item", {}))
+        if not isinstance(item, dict):
+            raise ValueError("mediaBinding.item must be an object")
+        item[ref_key] = visual_asset_id
+        payload = _build_custom_payload(ctx.template_config)
+        _write_payload(payload, [(path.split("."), [item])])
+        scene["customPayload"] = payload
+        del scene["visuals"]
+    return scene
+
+
 def _payload_scene(ctx: _SkeletonContext, scene_id: str, plan: dict) -> dict:
 
 
@@ -2979,7 +3002,7 @@ Examples:
         "--motion", default=None,
         help="Camera motion for every scene (animationHints.motion): static / kenburns-in / kenburns-out / "
              "pan-left / pan-right / float. Replaces the skeleton's alternating default. `auto` removes the "
-             "motion from every scene so the template picks (image-slide rotates push / pan / pull / pan).",
+             "motion from every scene so the template picks its own motion.",
     )
     parser.add_argument("--allow-digital-human", action="store_true", help="Allow digital-human assets")
     parser.add_argument("--avatar-id", default=None, help="Digital-human avatar id (dh_xxx) for templates declaring capabilities.digitalHuman: adds one gen-digital-human asset referenced by every scene's visuals.avatar.assetRef (one continuous presenter driven by the narration). Placement stays the template default unless set with --payload digitalHuman.<field>=...")

package/skills/gen-script/skill.json

@@ -67,7 +67,7 @@
       "narration": {
         "type": "string",
         "enum": ["auto", "on", "off"],
-        "description": "Narration on / off, overriding the template's needsNarration (auto = the template decides). Templates where narration is optional (image sequences, galton-board) default to none — **must** pass `on` when the replicate card says `旁白: <名称>（narrated）`, otherwise the narration you write is dropped without an error and the video comes back silent. `off` drops narration from a narrated template."
+        "description": "Narration on / off, overriding the template's needsNarration (auto = the template decides). Templates declaring narration optional default to none — **must** pass `on` when the replicate card says `旁白: <名称>（narrated）`, otherwise the narration you write is dropped without an error and the video comes back silent. `off` drops narration from a narrated template."
       },
       "background_asset": {
         "type": "string",

package/skills/gen-script/version.json

@@ -2,6 +2,6 @@
   "skillName": "gen-script",
   "repoName": "agent-skill-media-maker",
   "skillId": "474",
-  "version": "V25",
+  "version": "V26",
   "skillDescription": "Video-script generation skill. Turns a user-supplied topic into a structured Video DSL (JSON) that describes the full video — scene structure, asset requirements, and narrative flow.\n\nUse this skill as soon as the user mentions any of these intents:\n- Write a video script, generate a video script, plan a video, write storyboards\n- Create a short video, plan video content, generate a Video DSL\n- Turn a topic into a video structure / video content plan\n\nEven when the user does not say \"generate the DSL\", use this skill whenever they want to turn a topic into a structured video plan.\n\n⚠️ Stop-and-confirm gate: after this skill returns a DSL, show the full script and wait for the user's explicit confirmation. Never call `prepare_video_assets` in the same turn."
 }

package/skills/gen-voice/SKILL.md

@@ -1,23 +1,23 @@
 ---
 name: gen-voice
 description: |
-  Text-to-speech (TTS) skill: synthesize narration audio from text via ab-api's unified `/tool/tts` endpoint with `provider: "minimax"` (Minimax TTS).
+  Text-to-speech (TTS) skill: synthesize narration audio from text via ab-api's unified `/tool/tts` endpoint. Voices come from several providers (MiniMax, 豆包/Volcengine, HeyGen, plus the caller's cloned voices); the backend routes each voice id to its own provider.
 
   Use this skill immediately whenever the user asks for any of:
   - AI voice-over, TTS, text-to-speech, generate narration audio
-  - Use Minimax for speech synthesis
+  - Use MiniMax / 豆包 (Doubao) / HeyGen voices for speech synthesis
   - Read a piece of text aloud / produce an audio file from text
 
   Even when the user does not explicitly say "AI", any request that turns text into speech should route here.
 triggers:
   - AI voice-over, TTS, text-to-speech, generate narration audio
-  - Use Minimax for speech synthesis
+  - Use MiniMax / 豆包 (Doubao) / HeyGen voices for speech synthesis
   - Read a piece of text aloud / produce an audio file from text
 ---
 
 # Text-to-Speech (TTS) Skill
 
-Wraps ab-api's unified `POST /tool/tts` with `provider: "minimax"` (the same endpoint the web studio uses), authenticated with the **Tianyan privateToken**, backed by the **Minimax TTS** model.
+Wraps ab-api's unified `POST /tool/tts` (the same endpoint the web studio uses), authenticated with the **Tianyan privateToken**. You never pick a provider: ab-api looks the voice id up and synthesizes with that voice's provider (MiniMax, 豆包/Volcengine, HeyGen). Cloned voices (豆包 voice clones, HeyGen digital-twin voices) are usable only by their owner.
 
 ## Auth & environment
 
@@ -37,16 +37,21 @@ There is no skill-local env file — the executing process inherits the system e
 > This skill was migrated from a Python script to an remixmate CLI HTTP handler (`entry.type: http`). The agent invocation is unchanged (same tool name `gen_voice`, same params as in `skill.json`); local repro goes through `remixmate gen-voice ...`. The legacy `--download` flag has been removed — audio URLs are persisted in the cloud and play directly.
 
 1. **Text**: confirm what to synthesize. Punctuation drives pauses (commas short, periods long).
-2. **Voice**: if the user prefers a specific voice, run `--list-voices` first and pick a matching id.
+2. **Voice**: if the user prefers a specific voice, run `--list-voices` (with filters) first and pick a matching id.
 3. **Output**: by default the audio URL is printed; add `--json-output` for a structured result that includes per-line subtitle timestamps.
 
 ### List available voices
 
 ```bash
-remixmate gen-voice --list-voices
+remixmate gen-voice --list-voices                                   # page 1 of the public catalog
+remixmate gen-voice --list-voices --provider volcengine --gender female --language 中文-普通话
+remixmate gen-voice --list-voices --page 2                           # next 100
+remixmate gen-voice --list-voices --mine                             # your own cloned voices
 ```
 
-The list above is the **authoritative** catalog — a voice id is invalid only if it is missing from it.
+Each row shows the voice id, provider, type (system / clone), gender, language and name; the footer gives the total and the next `--page` when there is more. The public catalog has a few hundred voices, so filter before paging. `--json-output` returns `{page, total, voices:[{id,name,provider,type,gender,language}]}`.
+
+The live list is the **authoritative** catalog — a voice id is invalid only if it is missing from it (check the filters and later pages before concluding that).
 
 For offline development (no token / no network) there is a `--local` escape hatch. It prints the resolver's two hardcoded fallback ids, **not** the available voices, so never use it to decide whether a voice exists:
 
@@ -83,9 +88,16 @@ remixmate gen-voice --text "<text-to-synthesize>" --json-output
 |------|-------------|---------|
 | `--text` | Text to synthesize (required, except with `--list-voices`) | — |
 | `--voice-id` | Voice id (use `--list-voices` to discover) | `Chinese (Mandarin)_Male_Announcer` |
-| `--speed` | Speech rate, 0.5–2.0 | `1.0` |
-| `--language-boost` | CLI-only. Language hint for MiniMax (`auto` / `Chinese` / `English` / …). Mixed-script narration ("只有四个字：Agent Loop") mispronounces the Latin terms without one, so the service already applies its own default — pass this only to pin a specific language. Usually set from the DSL (`global.narration.languageBoost`, scene-level overrides) rather than by hand. Not exposed to the agent — see the handler comment. | service default |
-| `--list-voices` | List available voices and exit | — |
+| `--speed` | Speech rate, 0.5–2.0 (some HeyGen cloned voices ignore it) | `1.0` |
+| `--language-boost` | CLI-only. Language hint (MiniMax voices only) (`auto` / `Chinese` / `English` / …). Mixed-script narration ("只有四个字：Agent Loop") mispronounces the Latin terms without one, so the service already applies its own default — pass this only to pin a specific language. Usually set from the DSL (`global.narration.languageBoost`, scene-level overrides) rather than by hand. Not exposed to the agent — see the handler comment. | service default |
+| `--list-voices` | List voices and exit (100 per page) | — |
+| `--mine` | With `--list-voices`: your own cloned voices instead of the public catalog | off |
+| `--provider` | With `--list-voices`: `minimax` / `volcengine` (豆包) / `heygen` | all |
+| `--voice-type` | With `--list-voices`: `system` / `clone` | all |
+| `--gender` | With `--list-voices`: `male` / `female` | all |
+| `--language` | With `--list-voices`: exact language tag, e.g. `中文-普通话`, `英语` | all |
+| `--name` | With `--list-voices`: fuzzy name filter | — |
+| `--page` | With `--list-voices`: page number | `1` |
 | `--local` | CLI-only. With `--list-voices`: print the 2-entry hardcoded fallback list instead of querying the service (offline dev). Not exposed to the agent — see the handler comment. | off |
 | `--json-output` | Emit JSON (url, audio_length_ms, subtitles) | off |
 | `--priv-token` | Override token | env var |

package/skills/gen-voice/skill.json

@@ -3,8 +3,8 @@
   "toolName": "gen_voice",
   "tier": "atomic",
   "category": "asset",
-  "title": "Text-to-Speech (Minimax)",
-  "description": "Text-to-speech (TTS): synthesize narration audio from text via the Minimax TTS model. Returns the persisted audio URL — no download needed.",
+  "title": "Text-to-Speech",
+  "description": "Text-to-speech (TTS): synthesize narration audio from text. The voice catalog spans several providers (MiniMax, 豆包/Volcengine, HeyGen, and the caller's own cloned voices); the provider is picked from the voice id automatically. Returns the persisted audio URL — no download needed.",
   "auth": "required",
   "joinsTake": true,
   "envVars": [
@@ -25,15 +25,57 @@
       },
       "voice_id": {
         "type": "string",
-        "description": "Voice id. Default 'Chinese (Mandarin)_Male_Announcer'. When unsure, call with list_voices=true first to see what's available — do not invent ids, and do not declare a voice unavailable unless it is missing from that list."
+        "description": "Voice id from list_voices (any provider; synthesis routes to the voice's own provider). Default 'Chinese (Mandarin)_Male_Announcer'. When unsure, call with list_voices=true first — do not invent ids, and do not declare a voice unavailable unless it is missing from the filtered list."
       },
       "speed": {
         "type": "number",
-        "description": "Speech rate, 0.5-2.0"
+        "description": "Speech rate, 0.5-2.0. Some HeyGen cloned voices ignore it."
       },
       "list_voices": {
         "type": "boolean",
-        "description": "List the available voices and exit. This queries the live voice service and is the authoritative catalog — one call is enough, and its absence from this list is the only evidence that a voice id is invalid."
+        "description": "List voices and exit (one page of up to 100, with provider / type / gender / language columns and the total). This queries the live voice service and is the authoritative catalog. The public catalog has a few hundred voices, so narrow it with provider / gender / language / name, or page through with page; mine=true lists the caller's own cloned voices instead."
+      },
+      "mine": {
+        "type": "boolean",
+        "description": "With list_voices=true: list the caller's own cloned voices instead of the public catalog (provider / voice_type do not apply)."
+      },
+      "provider": {
+        "type": "string",
+        "enum": [
+          "minimax",
+          "volcengine",
+          "heygen"
+        ],
+        "description": "With list_voices=true: only voices from this provider (volcengine = 豆包). Not needed for synthesis."
+      },
+      "voice_type": {
+        "type": "string",
+        "enum": [
+          "system",
+          "clone"
+        ],
+        "description": "With list_voices=true: system (stock) or clone voices."
+      },
+      "gender": {
+        "type": "string",
+        "enum": [
+          "male",
+          "female"
+        ],
+        "description": "With list_voices=true: filter by gender."
+      },
+      "language": {
+        "type": "string",
+        "description": "With list_voices=true: filter by language tag, exact tag as shown in the Language column (e.g. 中文-普通话, 中文-粤语, 英语, 日语)."
+      },
+      "name": {
+        "type": "string",
+        "description": "With list_voices=true: fuzzy-filter by voice name."
+      },
+      "page": {
+        "type": "integer",
+        "minimum": 1,
+        "description": "With list_voices=true: page number (100 per page). Default 1."
       },
       "json_output": {
         "type": "boolean",
@@ -52,7 +94,14 @@
     ],
     "hidden": [
       "json_output",
-      "list_voices"
+      "list_voices",
+      "mine",
+      "provider",
+      "voice_type",
+      "gender",
+      "language",
+      "name",
+      "page"
     ]
   }
 }

package/skills/template-registry/version.json

@@ -2,6 +2,6 @@
   "skillName": "template-registry",
   "repoName": "agent-skill-media-maker",
   "skillId": "475",
-  "version": "V31",
+  "version": "V32",
   "skillDescription": "Video-template registry skill. Stores every video-template definition and lists the available templates (templateId / name / aspect ratio / style tags).\n\nUse this skill as soon as the user mentions any of these intents:\n- View available templates / list every template\n\nNote: DSL→TemplateBinding is no longer a separate exposed step — once prepare_video_assets receives a template_id it builds the binding internally."
 }

package/skills/template-registry/video_dsl/runtime/dsl_validator.py

@@ -1273,6 +1273,44 @@ def _check_narration_language_strict(dsl: dict) -> list[ValidationError]:
 _TEXT_LAYER_SKELETON_PREFIXES = ("[骨架待填充]", "[skeleton placeholder]")
 
 
+def _check_presenter_visibility(dsl: dict) -> list[ValidationError]:
+
+
+
+
+    tpl = _get_template_cfg(_pick_template_id(dsl)) or {}
+    caps = tpl.get("capabilities")
+    caps = caps if isinstance(caps, dict) else {}
+    spec = caps.get("digitalHuman") or {}
+    if not isinstance(spec, dict) or spec.get("backgroundOccludes") is not True:
+        return []
+    presenter_scenes = []
+    for scene in dsl.get("scenes") or []:
+        if not isinstance(scene, dict):
+            continue
+        visuals = scene.get("visuals") or {}
+        if not isinstance(visuals, dict):
+            continue
+        if not isinstance(visuals.get("avatar"), dict):
+            continue
+        if (visuals.get("avatar") or {}).get("assetRef"):
+            presenter_scenes.append(scene)
+    if not presenter_scenes or not all(
+        isinstance(scene["visuals"].get("background"), dict)
+        and scene["visuals"]["background"].get("assetRef")
+        for scene in presenter_scenes
+    ):
+        return []
+    binding = caps.get("mediaBinding") or {}
+    path = binding.get("itemsPath") if isinstance(binding, dict) else None
+    hint = f" Use timed media items in customPayload.{path} instead." if path else " Leave presenter scenes without full-scene backgrounds."
+    return [ValidationError(
+        "scenes[].visuals.background.assetRef",
+        "Full-scene backgrounds cover every presenter scene; the presenter would never be visible."
+        + hint,
+    )]
+
+
 def _check_text_layer_rules(dsl: dict) -> list[ValidationError]:
 
 
@@ -1398,6 +1436,7 @@ def validate_integrity(dsl: dict) -> list[ValidationError]:
     errors += _check_step_narration_alignment(dsl)
     errors += _check_slide_cross_field(dsl)
     errors += _check_text_layer_rules(dsl)
+    errors += _check_presenter_visibility(dsl)
 
     errors += _check_duration_estimate(dsl)
     return errors
@@ -1435,6 +1474,7 @@ def validate_dsl(dsl: dict) -> list[ValidationError]:
     errors += _check_step_narration_alignment(dsl)
     errors += _check_slide_cross_field(dsl)
     errors += _check_text_layer_rules(dsl)
+    errors += _check_presenter_visibility(dsl)
 
     errors += _check_duration_estimate(dsl)
     return errors