Update to Widemile
Snapshot Oct 29, 2026 · 00:05 UTC · version 1.2.0
Collection source: downloaded plugin package.
Payment or plan references changed
Skill instructions — Now says: the dedicated music preparation/execution pairs below, and `prostudio_job_status`. - **New version of an existing track:** extend, replace one section,...
Observed in published text. Live prices and checkout terms have not been verified by this change.
Up to 3 plugins free · no card required.
All fields and original values.
Skill instructions
Oct 24, 2026 · 13:07 UTC
`prostudio_music_edit_prepare`, `prostudio_music_edit`, and `prostudio_job_status`. - **New version of an existing track:** extend, replace one section, reinterpret, add vocals, add accompaniment, or separate stems. Use `prostudio_music_...
Oct 29, 2026 · 00:05 UTC
the dedicated music preparation/execution pairs below, and `prostudio_job_status`. - **New version of an existing track:** extend, replace one section, reinterpret, add vocals, add accompaniment, or separate stems. Use the matching dedic...
Compare saved observations
Download comparison JSONSKILL.md text changes
− removed · + added
--- before +++ after @@ -8,7 +8,7 @@ ## Tool requirements - **New composition:** `prostudio_models_list`, `prostudio_generation_prepare`, `prostudio_music_generate`, `prostudio_job_status`, and `prostudio_generation_get`. -- **Existing-track resolution and editing:** `prostudio_generations_list`, `prostudio_generation_get`, `prostudio_music_edit_prepare`, `prostudio_music_edit`, and `prostudio_job_status`. +- **Existing-track resolution and editing:** `prostudio_generations_list`, `prostudio_generation_get`, the dedicated music preparation/execution pairs below, and `prostudio_job_status`. - This declaration does not activate tools. Use only the tools required by the selected branch. If a required tool is absent from the current scoped catalog and `load_prostudio_tools` is available, call it with the missing outcome before claiming the capability is unavailable; do not call it when the tool is already available. If retrieval still exposes no matching tool, report that limitation. Treat music as a standalone creative workflow. Do not require, create, or read a production Brief solely to make or edit music. Load `generation-planning` as well only when creating a new track. Ask one concise question only when a missing answer would materially change the music or source selection; otherwise preserve the user's language, make conservative creative choices, and continue. @@ -16,7 +16,7 @@ ## Classify the request - **New composition:** create a new instrumental or vocal track from a musical direction. Use the normal generation plan and `prostudio_music_generate` flow. -- **New version of an existing track:** extend, replace one section, reinterpret, add vocals, add accompaniment, or separate stems. Use `prostudio_music_edit_prepare` and `prostudio_music_edit`; never simulate one of these operations with a whole-track regeneration. +- **New version of an existing track:** extend, replace one section, reinterpret, add vocals, add accompaniment, or separate stems. Use the matching dedicated preparation/execution pair below; never simulate one of these operations with a whole-track regeneration. - **Dependent sequence:** when the user requests multiple transformations whose second step depends on the first, run them sequentially. Wait for the first version to complete, then use its new generation id as the source of the next prepared operation. Never prepare the whole chain against the original source. - Music creation and music editing are distinct from voice-over and sound-effect generation. Do not route a song, score, backing track, or musical stem request to an audio-design tool. @@ -56,49 +56,50 @@ ## Choose and prepare exactly one edit operation -Call `prostudio_music_edit_prepare` with the smallest complete `edit` object for one of the following outcomes. +Choose the matching pair: `prostudio_music_extend_prepare` / `prostudio_music_extend`, `prostudio_music_upload_extend_prepare` / `prostudio_music_upload_extend`, `prostudio_music_replace_section_prepare` / `prostudio_music_replace_section`, `prostudio_music_cover_prepare` / `prostudio_music_cover`, `prostudio_music_add_vocals_prepare` / `prostudio_music_add_vocals`, `prostudio_music_add_instrumental_prepare` / `prostudio_music_add_instrumental`, `prostudio_music_separate_vocals_prepare` / `prostudio_music_separate_vocals`, `prostudio_music_split_stems_prepare` / `prostudio_music_split_stems`, or `prostudio_music_isolate_stem_prepare` / `prostudio_music_isolate_stem`. + +Pass `workspaceId`, `sourceGenerationId`, optional `providerTrackId` and the creative fields directly. There is no `edit`, `operation` or `separationType` argument. Each tool fixes its operation. Use the execution tool and arguments returned by preparation; never reconstruct a generic call. ### Extend -- Use `operation=extend` when the user wants the track to continue beyond its current material. +- Use `prostudio_music_extend_prepare` when the user wants the track to continue beyond its current material. - If the user wants Suno to preserve the original character automatically, set `useOriginalParameters=true`. Do not send `continueAt`, custom style, continuation prompt, or advanced controls in that mode. Tell the user that Suno inherits the source settings and automatically chooses where to continue. - For a directed continuation, set `useOriginalParameters=false`, choose `continueAt` strictly inside the source duration, and provide both `style` and `prompt`. `style` describes the musical character of the new part; `prompt` describes what happens after the chosen point—development, dynamics, structure, vocals, and ending. Add `lyrics` only when exact new sung words are required. ### Replace one section -- Use `operation=replace-section` only for a generated Suno track retaining its provider identity. Set `infillStartSeconds` and `infillEndSeconds` to an interval inside the track that is at least 10 seconds and no more than half the source duration. +- Use `prostudio_music_replace_section_prepare` only for a generated Suno track retaining its provider identity. Set `infillStartSeconds` and `infillEndSeconds` to an interval inside the track that is at least 10 seconds and no more than half the source duration. - Put the desired musical character in `style` and the local change in `prompt`. For a lyrical replacement, preserve the user's wording and provide the revised lyrics in `lyrics`; prefer the complete revised lyric sheet when available so the regenerated interval keeps its textual context. - Do not use this operation for a global restyle, and do not regenerate the unaffected portion merely because the requested interval is difficult. ### Reinterpret the whole track -- Use `operation=cover` when the user wants a recognizable new interpretation of the complete source with a different genre, rhythm, instrumentation, or production aesthetic. +- Use `prostudio_music_cover_prepare` when the user wants a recognizable new interpretation of the complete source with a different genre, rhythm, instrumentation, or production aesthetic. - Put the new musical identity in `style`. Set `instrumental=true` for an instrumental reinterpretation. For a vocal reinterpretation, set `instrumental=false` and provide the exact new-version lyrics in `lyrics`. - Explain that this creates a new interpretation while retaining recognizable musical identity; do not promise an identical arrangement or a surgical local change. ### Add vocals -- Use `operation=add-vocals` for adding a sung performance to an instrumental or music source. Set `instrumental=false`. +- Use `prostudio_music_add_vocals_prepare` for adding a sung performance to an instrumental or music source. Set `instrumental=false`. - Put timbre, genre, and vocal character in `style`. Build `prompt` as one provider-ready block containing both a short `VOCAL DIRECTION:` section—intensity, dynamics, harmonies, phrasing, and development—and a `LYRICS:` section with the exact words to sing. Also preserve those exact words in `lyrics` for version metadata and later editing. - Never copy the source track's unrelated prompt or lyrics into the new vocal prompt. If the user supplied lyrics, preserve them; if lyrics are missing and the request does not authorize writing them, ask for them rather than fabricating a song. ### Add instrumental accompaniment -- Use `operation=add-instrumental` to build instrumental backing around an existing vocal or musical source. +- Use `prostudio_music_add_instrumental_prepare` to build instrumental backing around an existing vocal or musical source. - Put only the accompaniment specification in `style`: instruments to add, genre, rhythm, intensity, entry points, and role in the arrangement. Do not prefill `prompt` or `lyrics` with the source transcript or original lyrics, and do not describe this as a general restyle. ### Separate stems -- Use `operation=separate-vocals` with exactly one separation mode: - - `separate_vocal` for vocals plus instrumental; - - `split_stem` for all available stems; - - `split_stem_advanced` for one requested stem. -- For `split_stem_advanced`, send exactly one canonical `stemName`: `Lead Vocal`, `Backing Vocals`, `Drums`, `Bass`, `Guitar`, `Keyboard`, `Strings`, `Brass`, `Woodwinds`, `Percussion`, `Synth`, or `FX`. Map ordinary user wording such as piano to `Keyboard`, effects to `FX`, or lead voice to `Lead Vocal`; ask only when the intended group is genuinely ambiguous. +- Use `prostudio_music_separate_vocals_prepare` for vocals plus accompaniment, `prostudio_music_split_stems_prepare` for all available stems, or `prostudio_music_isolate_stem_prepare` for one requested stem. +- For stem isolation, send exactly one canonical `stemName`: `Lead Vocal`, `Backing Vocals`, `Drums`, `Bass`, `Guitar`, `Keyboard`, `Strings`, `Brass`, `Woodwinds`, `Percussion`, `Synth`, or `FX`. Map piano to `Keyboard`, effects to `FX`, or lead voice to `Lead Vocal`; ask only when the intended group is ambiguous. - Treat the completed output as one new stem collection/version containing multiple selectable tracks. Do not report each returned stem as an unrelated generation. When acting on, previewing, downloading, sharing, or deleting one stem later, use the currently selected foreground track and its stable provider identity. ## Approval, execution, and completion -- The prepare tool does not start work. Show the source track, operation, essential creative settings, and exact credit charge returned by `prostudio_music_edit_prepare` or the generation plan. Stop for the host's required explicit approval; never state a currency estimate. +- The prepare tool does not start work. Show the source track, operation, essential creative settings, and exact credit charge returned by the matching preparation tool or the generation plan. Stop for the host's required explicit approval; show the server-provided EUR amount for accounts with `billingDisplay.unit = "EUR"`, and credits otherwise. Never invent a currency conversion. - After approval, execute the exact returned arguments with `userApproved=true`. Do not silently change the source, operation, interval, model, settings, or credit value between preparation and execution. If the credit value changes, prepare again and request approval for the new exact amount. - Music work is asynchronous. Poll `prostudio_job_status` using the returned `jobId` until it is completed or failed. Preview only a completed durable result. Do not claim the track, version, or stems are ready while the job is queued or processing. - Report the operation actually started or completed, source track, new version relationship, and resulting track or stem collection. On failure, report the actionable provider or validation error without claiming that a version was created. + +- Billing display is controlled by the authenticated account. When `billingDisplay.unit` is `EUR`, present the returned euro charge and balances; otherwise present credits. Preserve prepared execution arguments and `expectedCredits` exactly, without displaying those machine values as prices.
Full technical diff · 1 changed fields
changed /skill_md_contents
"---\nname: music-workflows\ndescription: Compose production-ready Suno V6 music with intentional structure, lyrics, duration, model choice, and advanced controls, or create precise new versions through extension, section replacement, reinterpretation, added vocals, added accompaniment, and stem separation.\n---\n\n# Music workflows\n\n## Tool requirements\n\n- **New composition:** `prostudio_models_list`, `prostudio_generation_prepare`, `prostudio_music_generate`, `prostudio_job_status`, and `prostudio_generation_get`.\n- **Existing-track resolution and editing:** `prostudio_generations_list`, `prostudio_generation_get`, `prostudio_music_edit_prepare`, `prostudio_music_edit`, and `prostudio_job_status`.\n- This declaration does not activate tools. Use only the tools required by the selected branch. If a required tool is absent from the current scoped catalog and `load_prostudio_tools` is available, call it with the missing outcome before claiming the capability is unavailable; do not call it when the tool is already available. If retrieval still exposes no matching tool, report that limitation.\n\nTreat music as a standalone creative workflow. Do not require, create, or read a production Brief solely to make or edit music. Load `generation-planning` as well only when creating a new track. Ask one concise question only when a missing answer would materially change the music or source selection; otherwise preserve the user's language, make conservative creative choices, and continue.\n\n## Classify the request\n\n- **New composition:** create a new instrumental or vocal track from a musical direction. Use the normal generation plan and `prostudio_music_generate` flow.\n- **New version of an existing track:** extend, replace one section, reinterpret, add vocals, add accompaniment, or separate stems. Use `prostudio_music_edit_prepare` and `prostudio_music_edit`; never simulate one of these operations with a whole-track regeneration.\n- **Dependent sequence:** when the user requests multiple transformations whose second step depends on the first, run them sequentially. Wait for the first version to complete, then use its new generation id as the source of the next prepared operation. Never prepare the whole chain against the original source.\n- Music creation and music editing are distinct from voice-over and sound-effect generation. Do not route a song, score, backing track, or musical stem request to an audio-design tool.\n\n## Gather only material creative choices\n\nBefore preparing a new composition, determine from the request or ask only for the missing choices that materially affect the result:\n\n1. instrumental or vocal;\n2. musical purpose and desired duration;\n3. genre, mood, energy, tempo or rhythmic feel, principal instrumentation, and structural arc;\n4. for vocals, the exact lyrics and any explicit vocal-timbre preference.\n\nDo not force the user through a questionnaire and do not ask about advanced numeric controls by default. If the user delegates the creative direction, choose a coherent direction from their goal. Do not invent lyrics when the user asked to preserve supplied wording, and do not silently convert a vocal request into an instrumental.\n\n## Compose a new track\n\n- Call `prostudio_models_list` with `mediaType=music` before every new composition and select only a currently available model. An explicit user choice wins. Otherwise use:\n - **Suno V6** for the default high-control, production-ready result;\n - **Suno V6 Wild** when the user explicitly wants unusual, exploratory, surprising, or genre-bending choices;\n - **Suno V6 Mini** when speed or lower cost matters more than maximum fidelity.\n- Prepare with `prostudio_generation_prepare` using `mediaType=music`, `mode=music`, the selected `modelId`, and a concise `prompt` that states the track's creative purpose. Put the production specification in `options.style`; do not dump lyrics into the style field.\n- Write `options.style` as a compact, concrete musical specification in this order when known: genre and era, tempo or rhythmic feel, key musical character, principal instruments, vocal character, energy and dynamics, structure or progression, mix and production aesthetic. Use descriptive musical language instead of artist imitation. Do not add requirements the user did not request merely to make the field longer.\n- Use `options.instrumental=true` and omit `options.lyrics` for an instrumental. For a vocal song, use `options.instrumental=false` and provide the exact intended words in `options.lyrics`. Structure longer lyrics with useful section markers such as `[Verse]`, `[Pre-Chorus]`, `[Chorus]`, `[Bridge]`, and `[Outro]`; keep stage directions separate from words that should be sung.\n- Put a meaningful library name in `options.title` when the user supplied one or a concise title follows naturally from the request. Do not treat the title as part of the style prompt.\n- When the user requests a duration, pass the exact integer as `options.durationSeconds`; the supported range is 10–480 seconds. Describe it as the requested generation duration, not as a guaranteed sample-accurate final runtime. If duration is absent and does not block the creative goal, let the provider use its default instead of interrogating the user or inventing false precision.\n- Use `options.negativeTags` only for sounds, instruments, vocal traits, genres, or production qualities the user actually wants excluded. Phrase exclusions as a short comma-separated set, not as a second positive prompt.\n- Set `options.vocalGender` only when the user explicitly requests a feminine (`f`) or masculine (`m`) vocal timbre. Never present it as a specific singer identity.\n- `styleWeight`, `weirdnessConstraint`, and `audioWeight` are values from 0 to 1. Omit them to retain product defaults unless the user asks for finer control or their direction makes the intent unambiguous. Higher style weight means closer adherence to the style description; higher weirdness means more unusual choices; higher audio weight means stronger source-track influence in supported source-based operations. Avoid stacking extreme values or promising deterministic results.\n- Keep generation prompts in English while preserving lyrics, quoted words, titles, proper names, and other exact content in the user's requested language.\n\n## Select the exact source for an edit\n\n- Every edit starts from one completed music `sourceGenerationId` in the current workspace. When necessary, use `prostudio_generations_list` with the music category and `prostudio_generation_get` to inspect candidates. Prefer the track the user directly referenced or the unambiguous active/recent result; if two plausible tracks remain, ask one concise source-selection question rather than guessing.\n- A generation can contain multiple provider tracks, especially after stem separation. When the user selects one of them, pass its stable returned `providerTrackId`; never invent a track id or infer one from a URL, filename, or display order.\n- Generative edits require a source created with Suno V6, V6 Wild, or V6 Mini. Stem separation remains available for other completed music sources. Imported audio is automatically routed through upload-and-extend when appropriate; do not expose `upload-extend` as a separate creative choice.\n- Reuse the source model unless the operation explicitly supports and the user requests another available V6-family model. Do not describe edit operations as selectable generation models.\n\n## Choose and prepare exactly one edit operation\n\nCall `prostudio_music_edit_prepare` with the smallest complete `edit` object for one of the following outcomes.\n\n### Extend\n\n- Use `operation=extend` when the user wants the track to continue beyond its current material.\n- If the user wants Suno to preserve the original character automatically, set `useOriginalParameters=true`. Do not send `continueAt`, custom style, continuation prompt, or advanced controls in that mode. Tell the user that Suno inherits the source settings and automatically chooses where to continue.\n- For a directed continuation, set `useOriginalParameters=false`, choose `continueAt` strictly inside the source duration, and provide both `style` and `prompt`. `style` describes the musical character of the new part; `prompt` describes what happens after the chosen point—development, dynamics, structure, vocals, and ending. Add `lyrics` only when exact new sung words are required.\n\n### Replace one section\n\n- Use `operation=replace-section` only for a generated Suno track retaining its provider identity. Set `infillStartSeconds` and `infillEndSeconds` to an interval inside the track that is at least 10 seconds and no more than half the source duration.\n- Put the desired musical character in `style` and the local change in `prompt`. For a lyrical replacement, preserve the user's wording and provide the revised lyrics in `lyrics`; prefer the complete revised lyric sheet when available so the regenerated interval keeps its textual context.\n- Do not use this operation for a global restyle, and do not regenerate the unaffected portion merely because the requested interval is difficult.\n\n### Reinterpret the whole track\n\n- Use `operation=cover` when the user wants a recognizable new interpretation of the complete source with a different genre, rhythm, instrumentation, or production aesthetic.\n- Put the new musical identity in `style`. Set `instrumental=true` for an instrumental reinterpretation. For a vocal reinterpretation, set `instrumental=false` and provide the exact new-version lyrics in `lyrics`.\n- Explain that this creates a new interpretation while retaining recognizable musical identity; do not promise an identical arrangement or a surgical local change.\n\n### Add vocals\n\n- Use `operation=add-vocals` for adding a sung performance to an instrumental or music source. Set `instrumental=false`.\n- Put timbre, genre, and vocal character in `style`. Build `prompt` as one provider-ready block containing both a short `VOCAL DIRECTION:` section—intensity, dynamics, harmonies, phrasing, and development—and a `LYRICS:` section with the exact words to sing. Also preserve those exact words in `lyrics` for version metadata and later editing.\n- Never copy the source track's unrelated prompt or lyrics into the new vocal prompt. If the user supplied lyrics, preserve them; if lyrics are missing and the request does not authorize writing them, ask for them rather than fabricating a song.\n\n### Add instrumental accompaniment\n\n- Use `operation=add-instrumental` to build instrumental backing around an existing vocal or musical source.\n- Put only the accompaniment specification in `style`: instruments to add, genre, rhythm, intensity, entry points, and role in the arrangement. Do not prefill `prompt` or `lyrics` with the source transcript or original lyrics, and do not describe this as a general restyle.\n\n### Separate stems\n\n- Use `operation=separate-vocals` with exactly one separation mode:\n - `separate_vocal` for vocals plus instrumental;\n - `split_stem` for all available stems;\n - `split_stem_advanced` for one requested stem.\n- For `split_stem_advanced`, send exactly one canonical `stemName`: `Lead Vocal`, `Backing Vocals`, `Drums`, `Bass`, `Guitar`, `Keyboard`, `Strings`, `Brass`, `Woodwinds`, `Percussion`, `Synth`, or `FX`. Map ordinary user wording such as piano to `Keyboard`, effects to `FX`, or lead voice to `Lead Vocal`; ask only when the intended group is genuinely ambiguous.\n- Treat the completed output as one new stem collection/version containing multiple selectable tracks. Do not report each returned stem as an unrelated generation. When acting on, previewing, downloading, sharing, or deleting one stem later, use the currently selected foreground track and its stable provider identity.\n\n## Approval, execution, and completion\n\n- The prepare tool does not start work. Show the source track, operation, essential creative settings, and exact credit charge returned by `prostudio_music_edit_prepare` or the generation plan. Stop for the host's required explicit approval; never state a currency estimate.\n- After approval, execute the exact returned arguments with `userApproved=true`. Do not silently change the source, operation, interval, model, settings, or credit value between preparation and execution. If the credit value changes, prepare again and request approval for the new exact amount.\n- Music work is asynchronous. Poll `prostudio_job_status` using the returned `jobId` until it is completed or failed. Preview only a completed durable result. Do not claim the track, version, or stems are ready while the job is queued or processing.\n- Report the operation actually started or completed, source track, new version relationship, and resulting track or stem collection. On failure, report the actionable provider or validation error without claiming that a version was created.\n"
"---\nname: music-workflows\ndescription: Compose production-ready Suno V6 music with intentional structure, lyrics, duration, model choice, and advanced controls, or create precise new versions through extension, section replacement, reinterpretation, added vocals, added accompaniment, and stem separation.\n---\n\n# Music workflows\n\n## Tool requirements\n\n- **New composition:** `prostudio_models_list`, `prostudio_generation_prepare`, `prostudio_music_generate`, `prostudio_job_status`, and `prostudio_generation_get`.\n- **Existing-track resolution and editing:** `prostudio_generations_list`, `prostudio_generation_get`, the dedicated music preparation/execution pairs below, and `prostudio_job_status`.\n- This declaration does not activate tools. Use only the tools required by the selected branch. If a required tool is absent from the current scoped catalog and `load_prostudio_tools` is available, call it with the missing outcome before claiming the capability is unavailable; do not call it when the tool is already available. If retrieval still exposes no matching tool, report that limitation.\n\nTreat music as a standalone creative workflow. Do not require, create, or read a production Brief solely to make or edit music. Load `generation-planning` as well only when creating a new track. Ask one concise question only when a missing answer would materially change the music or source selection; otherwise preserve the user's language, make conservative creative choices, and continue.\n\n## Classify the request\n\n- **New composition:** create a new instrumental or vocal track from a musical direction. Use the normal generation plan and `prostudio_music_generate` flow.\n- **New version of an existing track:** extend, replace one section, reinterpret, add vocals, add accompaniment, or separate stems. Use the matching dedicated preparation/execution pair below; never simulate one of these operations with a whole-track regeneration.\n- **Dependent sequence:** when the user requests multiple transformations whose second step depends on the first, run them sequentially. Wait for the first version to complete, then use its new generation id as the source of the next prepared operation. Never prepare the whole chain against the original source.\n- Music creation and music editing are distinct from voice-over and sound-effect generation. Do not route a song, score, backing track, or musical stem request to an audio-design tool.\n\n## Gather only material creative choices\n\nBefore preparing a new composition, determine from the request or ask only for the missing choices that materially affect the result:\n\n1. instrumental or vocal;\n2. musical purpose and desired duration;\n3. genre, mood, energy, tempo or rhythmic feel, principal instrumentation, and structural arc;\n4. for vocals, the exact lyrics and any explicit vocal-timbre preference.\n\nDo not force the user through a questionnaire and do not ask about advanced numeric controls by default. If the user delegates the creative direction, choose a coherent direction from their goal. Do not invent lyrics when the user asked to preserve supplied wording, and do not silently convert a vocal request into an instrumental.\n\n## Compose a new track\n\n- Call `prostudio_models_list` with `mediaType=music` before every new composition and select only a currently available model. An explicit user choice wins. Otherwise use:\n - **Suno V6** for the default high-control, production-ready result;\n - **Suno V6 Wild** when the user explicitly wants unusual, exploratory, surprising, or genre-bending choices;\n - **Suno V6 Mini** when speed or lower cost matters more than maximum fidelity.\n- Prepare with `prostudio_generation_prepare` using `mediaType=music`, `mode=music`, the selected `modelId`, and a concise `prompt` that states the track's creative purpose. Put the production specification in `options.style`; do not dump lyrics into the style field.\n- Write `options.style` as a compact, concrete musical specification in this order when known: genre and era, tempo or rhythmic feel, key musical character, principal instruments, vocal character, energy and dynamics, structure or progression, mix and production aesthetic. Use descriptive musical language instead of artist imitation. Do not add requirements the user did not request merely to make the field longer.\n- Use `options.instrumental=true` and omit `options.lyrics` for an instrumental. For a vocal song, use `options.instrumental=false` and provide the exact intended words in `options.lyrics`. Structure longer lyrics with useful section markers such as `[Verse]`, `[Pre-Chorus]`, `[Chorus]`, `[Bridge]`, and `[Outro]`; keep stage directions separate from words that should be sung.\n- Put a meaningful library name in `options.title` when the user supplied one or a concise title follows naturally from the request. Do not treat the title as part of the style prompt.\n- When the user requests a duration, pass the exact integer as `options.durationSeconds`; the supported range is 10–480 seconds. Describe it as the requested generation duration, not as a guaranteed sample-accurate final runtime. If duration is absent and does not block the creative goal, let the provider use its default instead of interrogating the user or inventing false precision.\n- Use `options.negativeTags` only for sounds, instruments, vocal traits, genres, or production qualities the user actually wants excluded. Phrase exclusions as a short comma-separated set, not as a second positive prompt.\n- Set `options.vocalGender` only when the user explicitly requests a feminine (`f`) or masculine (`m`) vocal timbre. Never present it as a specific singer identity.\n- `styleWeight`, `weirdnessConstraint`, and `audioWeight` are values from 0 to 1. Omit them to retain product defaults unless the user asks for finer control or their direction makes the intent unambiguous. Higher style weight means closer adherence to the style description; higher weirdness means more unusual choices; higher audio weight means stronger source-track influence in supported source-based operations. Avoid stacking extreme values or promising deterministic results.\n- Keep generation prompts in English while preserving lyrics, quoted words, titles, proper names, and other exact content in the user's requested language.\n\n## Select the exact source for an edit\n\n- Every edit starts from one completed music `sourceGenerationId` in the current workspace. When necessary, use `prostudio_generations_list` with the music category and `prostudio_generation_get` to inspect candidates. Prefer the track the user directly referenced or the unambiguous active/recent result; if two plausible tracks remain, ask one concise source-selection question rather than guessing.\n- A generation can contain multiple provider tracks, especially after stem separation. When the user selects one of them, pass its stable returned `providerTrackId`; never invent a track id or infer one from a URL, filename, or display order.\n- Generative edits require a source created with Suno V6, V6 Wild, or V6 Mini. Stem separation remains available for other completed music sources. Imported audio is automatically routed through upload-and-extend when appropriate; do not expose `upload-extend` as a separate creative choice.\n- Reuse the source model unless the operation explicitly supports and the user requests another available V6-family model. Do not describe edit operations as selectable generation models.\n\n## Choose and prepare exactly one edit operation\n\nChoose the matching pair: `prostudio_music_extend_prepare` / `prostudio_music_extend`, `prostudio_music_upload_extend_prepare` / `prostudio_music_upload_extend`, `prostudio_music_replace_section_prepare` / `prostudio_music_replace_section`, `prostudio_music_cover_prepare` / `prostudio_music_cover`, `prostudio_music_add_vocals_prepare` / `prostudio_music_add_vocals`, `prostudio_music_add_instrumental_prepare` / `prostudio_music_add_instrumental`, `prostudio_music_separate_vocals_prepare` / `prostudio_music_separate_vocals`, `prostudio_music_split_stems_prepare` / `prostudio_music_split_stems`, or `prostudio_music_isolate_stem_prepare` / `prostudio_music_isolate_stem`.\n\nPass `workspaceId`, `sourceGenerationId`, optional `providerTrackId` and the creative fields directly. There is no `edit`, `operation` or `separationType` argument. Each tool fixes its operation. Use the execution tool and arguments returned by preparation; never reconstruct a generic call.\n\n### Extend\n\n- Use `prostudio_music_extend_prepare` when the user wants the track to continue beyond its current material.\n- If the user wants Suno to preserve the original character automatically, set `useOriginalParameters=true`. Do not send `continueAt`, custom style, continuation prompt, or advanced controls in that mode. Tell the user that Suno inherits the source settings and automatically chooses where to continue.\n- For a directed continuation, set `useOriginalParameters=false`, choose `continueAt` strictly inside the source duration, and provide both `style` and `prompt`. `style` describes the musical character of the new part; `prompt` describes what happens after the chosen point—development, dynamics, structure, vocals, and ending. Add `lyrics` only when exact new sung words are required.\n\n### Replace one section\n\n- Use `prostudio_music_replace_section_prepare` only for a generated Suno track retaining its provider identity. Set `infillStartSeconds` and `infillEndSeconds` to an interval inside the track that is at least 10 seconds and no more than half the source duration.\n- Put the desired musical character in `style` and the local change in `prompt`. For a lyrical replacement, preserve the user's wording and provide the revised lyrics in `lyrics`; prefer the complete revised lyric sheet when available so the regenerated interval keeps its textual context.\n- Do not use this operation for a global restyle, and do not regenerate the unaffected portion merely because the requested interval is difficult.\n\n### Reinterpret the whole track\n\n- Use `prostudio_music_cover_prepare` when the user wants a recognizable new interpretation of the complete source with a different genre, rhythm, instrumentation, or production aesthetic.\n- Put the new musical identity in `style`. Set `instrumental=true` for an instrumental reinterpretation. For a vocal reinterpretation, set `instrumental=false` and provide the exact new-version lyrics in `lyrics`.\n- Explain that this creates a new interpretation while retaining recognizable musical identity; do not promise an identical arrangement or a surgical local change.\n\n### Add vocals\n\n- Use `prostudio_music_add_vocals_prepare` for adding a sung performance to an instrumental or music source. Set `instrumental=false`.\n- Put timbre, genre, and vocal character in `style`. Build `prompt` as one provider-ready block containing both a short `VOCAL DIRECTION:` section—intensity, dynamics, harmonies, phrasing, and development—and a `LYRICS:` section with the exact words to sing. Also preserve those exact words in `lyrics` for version metadata and later editing.\n- Never copy the source track's unrelated prompt or lyrics into the new vocal prompt. If the user supplied lyrics, preserve them; if lyrics are missing and the request does not authorize writing them, ask for them rather than fabricating a song.\n\n### Add instrumental accompaniment\n\n- Use `prostudio_music_add_instrumental_prepare` to build instrumental backing around an existing vocal or musical source.\n- Put only the accompaniment specification in `style`: instruments to add, genre, rhythm, intensity, entry points, and role in the arrangement. Do not prefill `prompt` or `lyrics` with the source transcript or original lyrics, and do not describe this as a general restyle.\n\n### Separate stems\n\n- Use `prostudio_music_separate_vocals_prepare` for vocals plus accompaniment, `prostudio_music_split_stems_prepare` for all available stems, or `prostudio_music_isolate_stem_prepare` for one requested stem.\n- For stem isolation, send exactly one canonical `stemName`: `Lead Vocal`, `Backing Vocals`, `Drums`, `Bass`, `Guitar`, `Keyboard`, `Strings`, `Brass`, `Woodwinds`, `Percussion`, `Synth`, or `FX`. Map piano to `Keyboard`, effects to `FX`, or lead voice to `Lead Vocal`; ask only when the intended group is ambiguous.\n- Treat the completed output as one new stem collection/version containing multiple selectable tracks. Do not report each returned stem as an unrelated generation. When acting on, previewing, downloading, sharing, or deleting one stem later, use the currently selected foreground track and its stable provider identity.\n\n## Approval, execution, and completion\n\n- The prepare tool does not start work. Show the source track, operation, essential creative settings, and exact credit charge returned by the matching preparation tool or the generation plan. Stop for the host's required explicit approval; show the server-provided EUR amount for accounts with `billingDisplay.unit = \"EUR\"`, and credits otherwise. Never invent a currency conversion.\n- After approval, execute the exact returned arguments with `userApproved=true`. Do not silently change the source, operation, interval, model, settings, or credit value between preparation and execution. If the credit value changes, prepare again and request approval for the new exact amount.\n- Music work is asynchronous. Poll `prostudio_job_status` using the returned `jobId` until it is completed or failed. Preview only a completed durable result. Do not claim the track, version, or stems are ready while the job is queued or processing.\n- Report the operation actually started or completed, source track, new version relationship, and resulting track or stem collection. On failure, report the actionable provider or validation error without claiming that a version was created.\n\n- Billing display is controlled by the authenticated account. When `billingDisplay.unit` is `EUR`, present the returned euro charge and balances; otherwise present credits. Preserve prepared execution arguments and `expectedCredits` exactly, without displaying those machine values as prices.\n"
Full snapshot data
{
"description": "Compose production-ready Suno V6 music with intentional structure, lyrics, duration, model choice, and advanced controls, or create precise new versions through extension, section replacement, reinterpretation, added vocals, added accompaniment, and stem separation.",
"included_files": [],
"name": "music-workflows",
"skill_md_contents": "---\nname: music-workflows\ndescription: Compose production-ready Suno V6 music with intentional structure, lyrics, duration, model choice, and advanced controls, or create precise new versions through extension, section replacement, reinterpretation, added vocals, added accompaniment, and stem separation.\n---\n\n# Music workflows\n\n## Tool requirements\n\n- **New composition:** `prostudio_models_list`, `prostudio_generation_prepare`, `prostudio_music_generate`, `prostudio_job_status`, and `prostudio_generation_get`.\n- **Existing-track resolution and editing:** `prostudio_generations_list`, `prostudio_generation_get`, the dedicated music preparation/execution pairs below, and `prostudio_job_status`.\n- This declaration does not activate tools. Use only the tools required by the selected branch. If a required tool is absent from the current scoped catalog and `load_prostudio_tools` is available, call it with the missing outcome before claiming the capability is unavailable; do not call it when the tool is already available. If retrieval still exposes no matching tool, report that limitation.\n\nTreat music as a standalone creative workflow. Do not require, create, or read a production Brief solely to make or edit music. Load `generation-planning` as well only when creating a new track. Ask one concise question only when a missing answer would materially change the music or source selection; otherwise preserve the user's language, make conservative creative choices, and continue.\n\n## Classify the request\n\n- **New composition:** create a new instrumental or vocal track from a musical direction. Use the normal generation plan and `prostudio_music_generate` flow.\n- **New version of an existing track:** extend, replace one section, reinterpret, add vocals, add accompaniment, or separate stems. Use the matching dedicated preparation/execution pair below; never simulate one of these operations with a whole-track regeneration.\n- **Dependent sequence:** when the user requests multiple transformations whose second step depends on the first, run them sequentially. Wait for the first version to complete, then use its new generation id as the source of the next prepared operation. Never prepare the whole chain against the original source.\n- Music creation and music editing are distinct from voice-over and sound-effect generation. Do not route a song, score, backing track, or musical stem request to an audio-design tool.\n\n## Gather only material creative choices\n\nBefore preparing a new composition, determine from the request or ask only for the missing choices that materially affect the result:\n\n1. instrumental or vocal;\n2. musical purpose and desired duration;\n3. genre, mood, energy, tempo or rhythmic feel, principal instrumentation, and structural arc;\n4. for vocals, the exact lyrics and any explicit vocal-timbre preference.\n\nDo not force the user through a questionnaire and do not ask about advanced numeric controls by default. If the user delegates the creative direction, choose a coherent direction from their goal. Do not invent lyrics when the user asked to preserve supplied wording, and do not silently convert a vocal request into an instrumental.\n\n## Compose a new track\n\n- Call `prostudio_models_list` with `mediaType=music` before every new composition and select only a currently available model. An explicit user choice wins. Otherwise use:\n - **Suno V6** for the default high-control, production-ready result;\n - **Suno V6 Wild** when the user explicitly wants unusual, exploratory, surprising, or genre-bending choices;\n - **Suno V6 Mini** when speed or lower cost matters more than maximum fidelity.\n- Prepare with `prostudio_generation_prepare` using `mediaType=music`, `mode=music`, the selected `modelId`, and a concise `prompt` that states the track's creative purpose. Put the production specification in `options.style`; do not dump lyrics into the style field.\n- Write `options.style` as a compact, concrete musical specification in this order when known: genre and era, tempo or rhythmic feel, key musical character, principal instruments, vocal character, energy and dynamics, structure or progression, mix and production aesthetic. Use descriptive musical language instead of artist imitation. Do not add requirements the user did not request merely to make the field longer.\n- Use `options.instrumental=true` and omit `options.lyrics` for an instrumental. For a vocal song, use `options.instrumental=false` and provide the exact intended words in `options.lyrics`. Structure longer lyrics with useful section markers such as `[Verse]`, `[Pre-Chorus]`, `[Chorus]`, `[Bridge]`, and `[Outro]`; keep stage directions separate from words that should be sung.\n- Put a meaningful library name in `options.title` when the user supplied one or a concise title follows naturally from the request. Do not treat the title as part of the style prompt.\n- When the user requests a duration, pass the exact integer as `options.durationSeconds`; the supported range is 10–480 seconds. Describe it as the requested generation duration, not as a guaranteed sample-accurate final runtime. If duration is absent and does not block the creative goal, let the provider use its default instead of interrogating the user or inventing false precision.\n- Use `options.negativeTags` only for sounds, instruments, vocal traits, genres, or production qualities the user actually wants excluded. Phrase exclusions as a short comma-separated set, not as a second positive prompt.\n- Set `options.vocalGender` only when the user explicitly requests a feminine (`f`) or masculine (`m`) vocal timbre. Never present it as a specific singer identity.\n- `styleWeight`, `weirdnessConstraint`, and `audioWeight` are values from 0 to 1. Omit them to retain product defaults unless the user asks for finer control or their direction makes the intent unambiguous. Higher style weight means closer adherence to the style description; higher weirdness means more unusual choices; higher audio weight means stronger source-track influence in supported source-based operations. Avoid stacking extreme values or promising deterministic results.\n- Keep generation prompts in English while preserving lyrics, quoted words, titles, proper names, and other exact content in the user's requested language.\n\n## Select the exact source for an edit\n\n- Every edit starts from one completed music `sourceGenerationId` in the current workspace. When necessary, use `prostudio_generations_list` with the music category and `prostudio_generation_get` to inspect candidates. Prefer the track the user directly referenced or the unambiguous active/recent result; if two plausible tracks remain, ask one concise source-selection question rather than guessing.\n- A generation can contain multiple provider tracks, especially after stem separation. When the user selects one of them, pass its stable returned `providerTrackId`; never invent a track id or infer one from a URL, filename, or display order.\n- Generative edits require a source created with Suno V6, V6 Wild, or V6 Mini. Stem separation remains available for other completed music sources. Imported audio is automatically routed through upload-and-extend when appropriate; do not expose `upload-extend` as a separate creative choice.\n- Reuse the source model unless the operation explicitly supports and the user requests another available V6-family model. Do not describe edit operations as selectable generation models.\n\n## Choose and prepare exactly one edit operation\n\nChoose the matching pair: `prostudio_music_extend_prepare` / `prostudio_music_extend`, `prostudio_music_upload_extend_prepare` / `prostudio_music_upload_extend`, `prostudio_music_replace_section_prepare` / `prostudio_music_replace_section`, `prostudio_music_cover_prepare` / `prostudio_music_cover`, `prostudio_music_add_vocals_prepare` / `prostudio_music_add_vocals`, `prostudio_music_add_instrumental_prepare` / `prostudio_music_add_instrumental`, `prostudio_music_separate_vocals_prepare` / `prostudio_music_separate_vocals`, `prostudio_music_split_stems_prepare` / `prostudio_music_split_stems`, or `prostudio_music_isolate_stem_prepare` / `prostudio_music_isolate_stem`.\n\nPass `workspaceId`, `sourceGenerationId`, optional `providerTrackId` and the creative fields directly. There is no `edit`, `operation` or `separationType` argument. Each tool fixes its operation. Use the execution tool and arguments returned by preparation; never reconstruct a generic call.\n\n### Extend\n\n- Use `prostudio_music_extend_prepare` when the user wants the track to continue beyond its current material.\n- If the user wants Suno to preserve the original character automatically, set `useOriginalParameters=true`. Do not send `continueAt`, custom style, continuation prompt, or advanced controls in that mode. Tell the user that Suno inherits the source settings and automatically chooses where to continue.\n- For a directed continuation, set `useOriginalParameters=false`, choose `continueAt` strictly inside the source duration, and provide both `style` and `prompt`. `style` describes the musical character of the new part; `prompt` describes what happens after the chosen point—development, dynamics, structure, vocals, and ending. Add `lyrics` only when exact new sung words are required.\n\n### Replace one section\n\n- Use `prostudio_music_replace_section_prepare` only for a generated Suno track retaining its provider identity. Set `infillStartSeconds` and `infillEndSeconds` to an interval inside the track that is at least 10 seconds and no more than half the source duration.\n- Put the desired musical character in `style` and the local change in `prompt`. For a lyrical replacement, preserve the user's wording and provide the revised lyrics in `lyrics`; prefer the complete revised lyric sheet when available so the regenerated interval keeps its textual context.\n- Do not use this operation for a global restyle, and do not regenerate the unaffected portion merely because the requested interval is difficult.\n\n### Reinterpret the whole track\n\n- Use `prostudio_music_cover_prepare` when the user wants a recognizable new interpretation of the complete source with a different genre, rhythm, instrumentation, or production aesthetic.\n- Put the new musical identity in `style`. Set `instrumental=true` for an instrumental reinterpretation. For a vocal reinterpretation, set `instrumental=false` and provide the exact new-version lyrics in `lyrics`.\n- Explain that this creates a new interpretation while retaining recognizable musical identity; do not promise an identical arrangement or a surgical local change.\n\n### Add vocals\n\n- Use `prostudio_music_add_vocals_prepare` for adding a sung performance to an instrumental or music source. Set `instrumental=false`.\n- Put timbre, genre, and vocal character in `style`. Build `prompt` as one provider-ready block containing both a short `VOCAL DIRECTION:` section—intensity, dynamics, harmonies, phrasing, and development—and a `LYRICS:` section with the exact words to sing. Also preserve those exact words in `lyrics` for version metadata and later editing.\n- Never copy the source track's unrelated prompt or lyrics into the new vocal prompt. If the user supplied lyrics, preserve them; if lyrics are missing and the request does not authorize writing them, ask for them rather than fabricating a song.\n\n### Add instrumental accompaniment\n\n- Use `prostudio_music_add_instrumental_prepare` to build instrumental backing around an existing vocal or musical source.\n- Put only the accompaniment specification in `style`: instruments to add, genre, rhythm, intensity, entry points, and role in the arrangement. Do not prefill `prompt` or `lyrics` with the source transcript or original lyrics, and do not describe this as a general restyle.\n\n### Separate stems\n\n- Use `prostudio_music_separate_vocals_prepare` for vocals plus accompaniment, `prostudio_music_split_stems_prepare` for all available stems, or `prostudio_music_isolate_stem_prepare` for one requested stem.\n- For stem isolation, send exactly one canonical `stemName`: `Lead Vocal`, `Backing Vocals`, `Drums`, `Bass`, `Guitar`, `Keyboard`, `Strings`, `Brass`, `Woodwinds`, `Percussion`, `Synth`, or `FX`. Map piano to `Keyboard`, effects to `FX`, or lead voice to `Lead Vocal`; ask only when the intended group is ambiguous.\n- Treat the completed output as one new stem collection/version containing multiple selectable tracks. Do not report each returned stem as an unrelated generation. When acting on, previewing, downloading, sharing, or deleting one stem later, use the currently selected foreground track and its stable provider identity.\n\n## Approval, execution, and completion\n\n- The prepare tool does not start work. Show the source track, operation, essential creative settings, and exact credit charge returned by the matching preparation tool or the generation plan. Stop for the host's required explicit approval; show the server-provided EUR amount for accounts with `billingDisplay.unit = \"EUR\"`, and credits otherwise. Never invent a currency conversion.\n- After approval, execute the exact returned arguments with `userApproved=true`. Do not silently change the source, operation, interval, model, settings, or credit value between preparation and execution. If the credit value changes, prepare again and request approval for the new exact amount.\n- Music work is asynchronous. Poll `prostudio_job_status` using the returned `jobId` until it is completed or failed. Preview only a completed durable result. Do not claim the track, version, or stems are ready while the job is queued or processing.\n- Report the operation actually started or completed, source track, new version relationship, and resulting track or stem collection. On failure, report the actionable provider or validation error without claiming that a version was created.\n\n- Billing display is controlled by the authenticated account. When `billingDisplay.unit` is `EUR`, present the returned euro charge and balances; otherwise present credits. Preserve prepared execution arguments and `expectedCredits` exactly, without displaying those machine values as prices.\n"
}SHA-256 of public snapshot: dcf3d7638258a9b22ad81438fd13e847b3321d30803bae8975e5ed88cd76c3b8