[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7C45HT3EQZKX9VYPMS6VBCH":3,"skill-same:01M7C45HT3EQZKX9VYPMS6VBCH":64},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":44,"commitSha":45,"skillDirectory":46,"forksCount":13,"repositoryHtmlUrl":47,"repositoryHomepageUrl":6,"ownerHtmlUrl":48,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":49,"install":50,"resources":63},"01M7C45HT3EQZKX9VYPMS6VBCH","songwriting-and-ai-music",null,"Songwriting craft and Suno AI music prompts.","fengwhang\u002Fmercury","fengwhang","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F325040254?v=4","01M7C3K54Y41KVGPHEMKEP9X70","01M7C3K55NTZ4B36YP2JY1NMJ0",0,"MIT","https:\u002F\u002Fgithub.com\u002Ffengwhang\u002Fmercury\u002Fblob\u002Fmain\u002Fhermes\u002Fskills\u002Fcreative\u002Fsongwriting-and-ai-music\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: songwriting-and-ai-music\ndescription: \"Songwriting craft and Suno AI music prompts.\"\nversion: 1.0.0\nauthor: Teknium (teknium1), Hermes Agent\nlicense: MIT\ntags: [songwriting, music, suno, parody, lyrics, creative]\nplatforms: [linux, macos, windows]\ntriggers:\n  - writing a song\n  - song lyrics\n  - music prompt\n  - suno prompt\n  - parody song\n  - adapting a song\n  - AI music generation\n---\n\n# Songwriting & AI Music Generation\n\nEverything here is a GUIDELINE, not a rule. Art breaks rules on purpose.\nUse what serves the song. Ignore what doesn't.\n\n---\n\n## 1. Song Structure (Pick One or Invent Your Own)\n\nCommon skeletons — mix, modify, or throw out as needed:\n\n```\nABABCB  Verse\u002FChorus\u002FVerse\u002FChorus\u002FBridge\u002FChorus    (most pop\u002Frock)\nAABA    Verse\u002FVerse\u002FBridge\u002FVerse (refrain-based)    (jazz standards, ballads)\nABAB    Verse\u002FChorus alternating                    (simple, direct)\nAAA     Verse\u002FVerse\u002FVerse (strophic, no chorus)     (folk, storytelling)\n```\n\nThe six building blocks:\n- Intro      — set the mood, pull the listener in\n- Verse      — the story, the details, the world-building\n- Pre-Chorus — optional tension ramp before the payoff\n- Chorus     — the emotional core, the part people remember\n- Bridge     — a detour, a shift in perspective or key\n- Outro      — the farewell, can echo or subvert the rest\n\nYou don't need all of these. Some great songs are just one section\nthat evolves. Structure serves the emotion, not the other way around.\n\n---\n\n## 2. Rhyme, Meter, and Sound\n\nRHYME TYPES (from tight to loose):\n- Perfect: lean\u002Fmean\n- Family: crate\u002Fbraid\n- Assonance: had\u002Fglass (same vowels, different endings)\n- Consonance: scene\u002Fwhen (different vowels, similar endings)\n- Near\u002Fslant: enough to suggest connection without locking it down\n\nMix them. All perfect rhymes can sound like a nursery rhyme.\nAll slant rhymes can sound lazy. The blend is where it lives.\n\nINTERNAL RHYME: Rhyming within a line, not just at the ends.\n  \"We pruned the lies from bleeding trees \u002F Distilled the storm\n   from entropy\" — \"lies\u002Fflies,\" \"trees\u002Fentropy\" create internal echoes.\n\nMETER: The rhythm of stressed vs unstressed syllables.\n- Matching syllable counts between parallel lines helps singability\n- The STRESSED syllables matter more than total count\n- Say it out loud. If you stumble, the meter needs work.\n- Intentionally breaking meter can create emphasis or surprise\n\n---\n\n## 3. Emotional Arc and Dynamics\n\nThink of a song as a journey, not a flat road.\n\nENERGY MAPPING (rough idea, not prescription):\n  Intro: 2-3  |  Verse: 5-6  |  Pre-Chorus: 7\n  Chorus: 8-9  |  Bridge: varies  |  Final Chorus: 9-10\n\nThe most powerful dynamic trick: CONTRAST.\n- Whisper before a scream hits harder than just screaming\n- Sparse before dense. Slow before fast. Low before high.\n- The drop only works because of the buildup\n- Silence is an instrument\n\n\"Whisper to roar to whisper\" — start intimate, build to full power,\nstrip back to vulnerability. Works for ballads, epics, anthems.\n\n---\n\n## 4. Writing Lyrics That Work\n\nSHOW, DON'T TELL (usually):\n- \"I was sad\" = flat\n- \"Your hoodie's still on the hook by the door\" = alive\n- But sometimes \"I give my life\" said plainly IS the power\n\nTHE HOOK:\n- The line people remember, hum, repeat\n- Usually the title or core phrase\n- Works best when melody + lyric + emotion all align\n- Place it where it lands hardest (often first\u002Flast line of chorus)\n\nPROSODY — lyrics and music supporting each other:\n- Stable feelings (resolution, peace) pair with settled melodies,\n  perfect rhymes, resolved chords\n- Unstable feelings (longing, doubt) pair with wandering melodies,\n  near-rhymes, unresolved chords\n- Verse melody typically sits lower, chorus goes higher\n- But flip this if it serves the song\n\nAVOID (unless you're doing it on purpose):\n- Cliches on autopilot (\"heart of gold\" without earning it)\n- Forcing word order to hit a rhyme (\"Yoda-speak\")\n- Same energy in every section (flat dynamics)\n- Treating your first draft as sacred — revision is creation\n\n---\n\n## 5. Parody and Adaptation\n\nWhen rewriting an existing song with new lyrics:\n\nTHE SKELETON: Map the original's structure first.\n- Count syllables per line\n- Mark the rhyme scheme (ABAB, AABB, etc.)\n- Identify which syllables are STRESSED\n- Note where held\u002Fsustained notes fall\n\nFITTING NEW WORDS:\n- Match stressed syllables to the same beats as the original\n- Total syllable count can flex by 1-2 unstressed syllables\n- On long held notes, try to match the VOWEL SOUND of the original\n  (if original holds \"LOOOVE\" with an \"oo\" vowel, \"FOOOD\" fits\n   better than \"LIFE\")\n- Monosyllabic swaps in key spots keep rhythm intact\n  (Crime -> Code, Snake -> Noose)\n- Sing your new words over the original — if you stumble, revise\n\nCONCEPT:\n- Pick a concept strong enough to sustain the whole song\n- Start from the title\u002Fhook and build outward\n- Generate lots of raw material (puns, phrases, images) FIRST,\n  then fit the best ones into the structure\n- If you need a specific line somewhere, reverse-engineer the\n  rhyme scheme backward to set it up\n\nKEEP SOME ORIGINALS: Leaving a few original lines or structures\nintact adds recognizability and lets the audience feel the connection.\n\n---\n\n## 6. Suno AI Prompt Engineering\n\n### Style\u002FGenre Description Field\n\nFORMULA (adapt as needed):\n  Genre + Mood + Era + Instruments + Vocal Style + Production + Dynamics\n\n```\nBAD:  \"sad rock song\"\nGOOD: \"Cinematic orchestral spy thriller, 1960s Cold War era, smoky\n       sultry female vocalist, big band jazz, brass section with\n       trumpets and french horns, sweeping strings, minor key,\n       vintage analog warmth\"\n```\n\nDESCRIBE THE JOURNEY, not just the genre:\n```\n\"Begins as a haunting whisper over sparse piano. Gradually layers\n in muted brass. Builds through the chorus with full orchestra.\n Second verse erupts with raw belting intensity. Outro strips back\n to a lone piano and a fragile whisper fading to silence.\"\n```\n\nTIPS:\n- V4.5+ supports up to 1,000 chars in Style field — use them\n- NO artist names or trademarks. Describe the sound instead.\n  \"1960s Cold War spy thriller brass\" not \"James Bond style\"\n  \"90s grunge\" not \"Nirvana-style\"\n- Specify BPM and key when you have a preference\n- Use Exclude Styles field for what you DON'T want\n- Unexpected genre combos can be gold: \"bossa nova trap\",\n  \"Appalachian gothic\", \"chiptune jazz\"\n- Build a vocal PERSONA, not just a gender:\n  \"A weathered torch singer with a smoky alto, slight rasp,\n   who starts vulnerable and builds to devastating power\"\n\n### Metatags (place in [brackets] inside lyrics field)\n\nSTRUCTURE:\n  [Intro] [Verse] [Verse 1] [Pre-Chorus] [Chorus]\n  [Post-Chorus] [Hook] [Bridge] [Interlude]\n  [Instrumental] [Instrumental Break] [Guitar Solo]\n  [Breakdown] [Build-up] [Outro] [Silence] [End]\n\nVOCAL PERFORMANCE:\n  [Whispered] [Spoken Word] [Belted] [Falsetto] [Powerful]\n  [Soulful] [Raspy] [Breathy] [Smooth] [Gritty]\n  [Staccato] [Legato] [Vibrato] [Melismatic]\n  [Harmonies] [Choir] [Harmonized Chorus]\n\nDYNAMICS:\n  [High Energy] [Low Energy] [Building Energy] [Explosive]\n  [Emotional Climax] [Gradual swell] [Orchestral swell]\n  [Quiet arrangement] [Falling tension] [Slow Down]\n\nGENDER:\n  [Female Vocals] [Male Vocals]\n\nATMOSPHERE:\n  [Melancholic] [Euphoric] [Nostalgic] [Aggressive]\n  [Dreamy] [Intimate] [Dark Atmosphere]\n\nSFX:\n  [Vinyl Crackle] [Rain] [Applause] [Static] [Thunder]\n\nPut tags in BOTH style field AND lyrics for reinforcement.\nKeep to 5-8 tags per section max — too many confuses the AI.\nDon't contradict yourself ([Calm] + [Aggressive] in same section).\n\n### Custom Mode\n- Always use Custom Mode for serious work (separate Style + Lyrics)\n- Lyrics field limit: ~3,000 chars (~40-60 lines)\n- Always add structural tags — without them Suno defaults to\n  flat verse\u002Fchorus\u002Fverse with no emotional arc\n\n---\n\n## 7. Phonetic Tricks for AI Singers\n\nAI vocalists don't read — they pronounce. Help them:\n\nPHONETIC RESPELLING:\n- Spell words as they SOUND: \"through\" -> \"thru\"\n- Proper nouns are highest failure rate — test early\n- \"Nous\" -> \"Noose\" (forces correct pronunciation)\n- Hyphenate to guide syllables: \"Re-search\", \"bio-engineering\"\n\nDELIVERY CONTROL:\n- ALL CAPS = louder, more intense\n- Vowel extension: \"lo-o-o-ove\" = sustained\u002Fmelisma\n- Ellipses: \"I... need... you\" = dramatic pauses\n- Hyphenated stretch: \"ne-e-ed\" = emotional stretch\n\nALWAYS:\n- Spell out numbers: \"24\u002F7\" -> \"twenty four seven\"\n- Space acronyms: \"AI\" -> \"A I\" or \"A-I\"\n- Test proper nouns\u002Funusual words in a short 30-second clip first\n- Once generated, pronunciation is baked in — fix in lyrics BEFORE\n\n---\n\n## 8. Workflow\n\n1. Write the concept\u002Fhook first — what's the emotional core?\n2. If adapting, map the original structure (syllables, rhyme, stress)\n3. Generate raw material — brainstorm freely before structuring\n4. Draft lyrics into the structure\n5. Read\u002Fsing aloud — catch stumbles, fix meter\n6. Build the Suno style description — paint the dynamic journey\n7. Add metatags to lyrics for performance direction\n8. Generate 3-5 variations minimum — treat them like recording takes\n9. Pick the best, use Extend\u002FContinue to build on promising sections\n10. If something great happens by accident, keep it\n\nEXPECT: ~3-5 generations per 1 good result. Revision is normal.\nStyle can drift in extensions — restate genre\u002Fmood when extending.\n\n---\n\n## 9. Lessons Learned\n\n- Describing the dynamic ARC in the style field matters way more\n  than just listing genres. \"Whisper to roar to whisper\" gives\n  Suno a performance map.\n- Keeping some original lines intact in a parody adds recognizability\n  and emotional weight — the audience feels the ghost of the original.\n- The bridge slot in a song is where you can transform imagery.\n  Swap the original's specific references for your theme's metaphors\n  while keeping the emotional function (reflection, shift, revelation).\n- Monosyllabic word swaps in hooks\u002Ftags are the cleanest way to\n  maintain rhythm while changing meaning.\n- A strong vocal persona description in the style field makes a\n  bigger difference than any single metatag.\n- Don't be precious about rules. If a line breaks meter but hits\n  harder, keep it. The feeling is what matters. Craft serves art,\n  not the other way around.\n\n---\n\n## 10. Local \u002F Open-Source Music Generation\n\nFor local, GPU-based generation instead of Suno, two optional skills\ncover this (heavy dependencies, so not installed by default):\n\n- **heartmula** — full songs with vocals from lyrics + tags\n  (open-source Suno alternative, 8-16GB VRAM):\n  `mercury skills install official\u002Fcreative\u002Fheartmula`\n- **audiocraft** — Meta's MusicGen (instrumental text-to-music) and\n  AudioGen (sound effects):\n  `mercury skills install official\u002Fcreative\u002Faudiocraft-audio-generation`\n\nThe lyric-writing and prompting craft in this skill applies to\nheartmula too — its input format is lyrics with bracketed structure\ntags plus comma-separated style tags.\n","# Songwriting & AI Music Generation\n\nEverything here is a GUIDELINE, not a rule. Art breaks rules on purpose.\nUse what serves the song. Ignore what doesn't.\n\n---\n\n## 1. Song Structure (Pick One or Invent Your Own)\n\nCommon skeletons — mix, modify, or throw out as needed:\n\n```\nABABCB  Verse\u002FChorus\u002FVerse\u002FChorus\u002FBridge\u002FChorus    (most pop\u002Frock)\nAABA    Verse\u002FVerse\u002FBridge\u002FVerse (refrain-based)    (jazz standards, ballads)\nABAB    Verse\u002FChorus alternating                    (simple, direct)\nAAA     Verse\u002FVerse\u002FVerse (strophic, no chorus)     (folk, storytelling)\n```\n\nThe six building blocks:\n- Intro      — set the mood, pull the listener in\n- Verse      — the story, the details, the world-building\n- Pre-Chorus — optional tension ramp before the payoff\n- Chorus     — the emotional core, the part people remember\n- Bridge     — a detour, a shift in perspective or key\n- Outro      — the farewell, can echo or subvert the rest\n\nYou don't need all of these. Some great songs are just one section\nthat evolves. Structure serves the emotion, not the other way around.\n\n---\n\n## 2. Rhyme, Meter, and Sound\n\nRHYME TYPES (from tight to loose):\n- Perfect: lean\u002Fmean\n- Family: crate\u002Fbraid\n- Assonance: had\u002Fglass (same vowels, different endings)\n- Consonance: scene\u002Fwhen (different vowels, similar endings)\n- Near\u002Fslant: enough to suggest connection without locking it down\n\nMix them. All perfect rhymes can sound like a nursery rhyme.\nAll slant rhymes can sound lazy. The blend is where it lives.\n\nINTERNAL RHYME: Rhyming within a line, not just at the ends.\n  \"We pruned the lies from bleeding trees \u002F Distilled the storm\n   from entropy\" — \"lies\u002Fflies,\" \"trees\u002Fentropy\" create internal echoes.\n\nMETER: The rhythm of stressed vs unstressed syllables.\n- Matching syllable counts between parallel lines helps singability\n- The STRESSED syllables matter more than total count\n- Say it out loud. If you stumble, the meter needs work.\n- Intentionally breaking meter can create emphasis or surprise\n\n---\n\n## 3. Emotional Arc and Dynamics\n\nThink of a song as a journey, not a flat road.\n\nENERGY MAPPING (rough idea, not prescription):\n  Intro: 2-3  |  Verse: 5-6  |  Pre-Chorus: 7\n  Chorus: 8-9  |  Bridge: varies  |  Final Chorus: 9-10\n\nThe most powerful dynamic trick: CONTRAST.\n- Whisper before a scream hits harder than just screaming\n- Sparse before dense. Slow before fast. Low before high.\n- The drop only works because of the buildup\n- Silence is an instrument\n\n\"Whisper to roar to whisper\" — start intimate, build to full power,\nstrip back to vulnerability. Works for ballads, epics, anthems.\n\n---\n\n## 4. Writing Lyrics That Work\n\nSHOW, DON'T TELL (usually):\n- \"I was sad\" = flat\n- \"Your hoodie's still on the hook by the door\" = alive\n- But sometimes \"I give my life\" said plainly IS the power\n\nTHE HOOK:\n- The line people remember, hum, repeat\n- Usually the title or core phrase\n- Works best when melody + lyric + emotion all align\n- Place it where it lands hardest (often first\u002Flast line of chorus)\n\nPROSODY — lyrics and music supporting each other:\n- Stable feelings (resolution, peace) pair with settled melodies,\n  perfect rhymes, resolved chords\n- Unstable feelings (longing, doubt) pair with wandering melodies,\n  near-rhymes, unresolved chords\n- Verse melody typically sits lower, chorus goes higher\n- But flip this if it serves the song\n\nAVOID (unless you're doing it on purpose):\n- Cliches on autopilot (\"heart of gold\" without earning it)\n- Forcing word order to hit a rhyme (\"Yoda-speak\")\n- Same energy in every section (flat dynamics)\n- Treating your first draft as sacred — revision is creation\n\n---\n\n## 5. Parody and Adaptation\n\nWhen rewriting an existing song with new lyrics:\n\nTHE SKELETON: Map the original's structure first.\n- Count syllables per line\n- Mark the rhyme scheme (ABAB, AABB, etc.)\n- Identify which syllables are STRESSED\n- Note where held\u002Fsustained notes fall\n\nFITTING NEW WORDS:\n- Match stressed syllables to the same beats as the original\n- Total syllable count can flex by 1-2 unstressed syllables\n- On long held notes, try to match the VOWEL SOUND of the original\n  (if original holds \"LOOOVE\" with an \"oo\" vowel, \"FOOOD\" fits\n   better than \"LIFE\")\n- Monosyllabic swaps in key spots keep rhythm intact\n  (Crime -> Code, Snake -> Noose)\n- Sing your new words over the original — if you stumble, revise\n\nCONCEPT:\n- Pick a concept strong enough to sustain the whole song\n- Start from the title\u002Fhook and build outward\n- Generate lots of raw material (puns, phrases, images) FIRST,\n  then fit the best ones into the structure\n- If you need a specific line somewhere, reverse-engineer the\n  rhyme scheme backward to set it up\n\nKEEP SOME ORIGINALS: Leaving a few original lines or structures\nintact adds recognizability and lets the audience feel the connection.\n\n---\n\n## 6. Suno AI Prompt Engineering\n\n### Style\u002FGenre Description Field\n\nFORMULA (adapt as needed):\n  Genre + Mood + Era + Instruments + Vocal Style + Production + Dynamics\n\n```\nBAD:  \"sad rock song\"\nGOOD: \"Cinematic orchestral spy thriller, 1960s Cold War era, smoky\n       sultry female vocalist, big band jazz, brass section with\n       trumpets and french horns, sweeping strings, minor key,\n       vintage analog warmth\"\n```\n\nDESCRIBE THE JOURNEY, not just the genre:\n```\n\"Begins as a haunting whisper over sparse piano. Gradually layers\n in muted brass. Builds through the chorus with full orchestra.\n Second verse erupts with raw belting intensity. Outro strips back\n to a lone piano and a fragile whisper fading to silence.\"\n```\n\nTIPS:\n- V4.5+ supports up to 1,000 chars in Style field — use them\n- NO artist names or trademarks. Describe the sound instead.\n  \"1960s Cold War spy thriller brass\" not \"James Bond style\"\n  \"90s grunge\" not \"Nirvana-style\"\n- Specify BPM and key when you have a preference\n- Use Exclude Styles field for what you DON'T want\n- Unexpected genre combos can be gold: \"bossa nova trap\",\n  \"Appalachian gothic\", \"chiptune jazz\"\n- Build a vocal PERSONA, not just a gender:\n  \"A weathered torch singer with a smoky alto, slight rasp,\n   who starts vulnerable and builds to devastating power\"\n\n### Metatags (place in [brackets] inside lyrics field)\n\nSTRUCTURE:\n  [Intro] [Verse] [Verse 1] [Pre-Chorus] [Chorus]\n  [Post-Chorus] [Hook] [Bridge] [Interlude]\n  [Instrumental] [Instrumental Break] [Guitar Solo]\n  [Breakdown] [Build-up] [Outro] [Silence] [End]\n\nVOCAL PERFORMANCE:\n  [Whispered] [Spoken Word] [Belted] [Falsetto] [Powerful]\n  [Soulful] [Raspy] [Breathy] [Smooth] [Gritty]\n  [Staccato] [Legato] [Vibrato] [Melismatic]\n  [Harmonies] [Choir] [Harmonized Chorus]\n\nDYNAMICS:\n  [High Energy] [Low Energy] [Building Energy] [Explosive]\n  [Emotional Climax] [Gradual swell] [Orchestral swell]\n  [Quiet arrangement] [Falling tension] [Slow Down]\n\nGENDER:\n  [Female Vocals] [Male Vocals]\n\nATMOSPHERE:\n  [Melancholic] [Euphoric] [Nostalgic] [Aggressive]\n  [Dreamy] [Intimate] [Dark Atmosphere]\n\nSFX:\n  [Vinyl Crackle] [Rain] [Applause] [Static] [Thunder]\n\nPut tags in BOTH style field AND lyrics for reinforcement.\nKeep to 5-8 tags per section max — too many confuses the AI.\nDon't contradict yourself ([Calm] + [Aggressive] in same section).\n\n### Custom Mode\n- Always use Custom Mode for serious work (separate Style + Lyrics)\n- Lyrics field limit: ~3,000 chars (~40-60 lines)\n- Always add structural tags — without them Suno defaults to\n  flat verse\u002Fchorus\u002Fverse with no emotional arc\n\n---\n\n## 7. Phonetic Tricks for AI Singers\n\nAI vocalists don't read — they pronounce. Help them:\n\nPHONETIC RESPELLING:\n- Spell words as they SOUND: \"through\" -> \"thru\"\n- Proper nouns are highest failure rate — test early\n- \"Nous\" -> \"Noose\" (forces correct pronunciation)\n- Hyphenate to guide syllables: \"Re-search\", \"bio-engineering\"\n\nDELIVERY CONTROL:\n- ALL CAPS = louder, more intense\n- Vowel extension: \"lo-o-o-ove\" = sustained\u002Fmelisma\n- Ellipses: \"I... need... you\" = dramatic pauses\n- Hyphenated stretch: \"ne-e-ed\" = emotional stretch\n\nALWAYS:\n- Spell out numbers: \"24\u002F7\" -> \"twenty four seven\"\n- Space acronyms: \"AI\" -> \"A I\" or \"A-I\"\n- Test proper nouns\u002Funusual words in a short 30-second clip first\n- Once generated, pronunciation is baked in — fix in lyrics BEFORE\n\n---\n\n## 8. Workflow\n\n1. Write the concept\u002Fhook first — what's the emotional core?\n2. If adapting, map the original structure (syllables, rhyme, stress)\n3. Generate raw material — brainstorm freely before structuring\n4. Draft lyrics into the structure\n5. Read\u002Fsing aloud — catch stumbles, fix meter\n6. Build the Suno style description — paint the dynamic journey\n7. Add metatags to lyrics for performance direction\n8. Generate 3-5 variations minimum — treat them like recording takes\n9. Pick the best, use Extend\u002FContinue to build on promising sections\n10. If something great happens by accident, keep it\n\nEXPECT: ~3-5 generations per 1 good result. Revision is normal.\nStyle can drift in extensions — restate genre\u002Fmood when extending.\n\n---\n\n## 9. Lessons Learned\n\n- Describing the dynamic ARC in the style field matters way more\n  than just listing genres. \"Whisper to roar to whisper\" gives\n  Suno a performance map.\n- Keeping some original lines intact in a parody adds recognizability\n  and emotional weight — the audience feels the ghost of the original.\n- The bridge slot in a song is where you can transform imagery.\n  Swap the original's specific references for your theme's metaphors\n  while keeping the emotional function (reflection, shift, revelation).\n- Monosyllabic word swaps in hooks\u002Ftags are the cleanest way to\n  maintain rhythm while changing meaning.\n- A strong vocal persona description in the style field makes a\n  bigger difference than any single metatag.\n- Don't be precious about rules. If a line breaks meter but hits\n  harder, keep it. The feeling is what matters. Craft serves art,\n  not the other way around.\n\n---\n\n## 10. Local \u002F Open-Source Music Generation\n\nFor local, GPU-based generation instead of Suno, two optional skills\ncover this (heavy dependencies, so not installed by default):\n\n- **heartmula** — full songs with vocals from lyrics + tags\n  (open-source Suno alternative, 8-16GB VRAM):\n  `mercury skills install official\u002Fcreative\u002Fheartmula`\n- **audiocraft** — Meta's MusicGen (instrumental text-to-music) and\n  AudioGen (sound effects):\n  `mercury skills install official\u002Fcreative\u002Faudiocraft-audio-generation`\n\nThe lyric-writing and prompting craft in this skill applies to\nheartmula too — its input format is lyrics with bracketed structure\ntags plus comma-separated style tags.\n",{"name":5,"tags":23,"author":30,"license":14,"version":31,"triggers":32,"platforms":40,"description":7},[24,25,26,27,28,29],"songwriting","music","suno","parody","lyrics","creative","Teknium (teknium1), Hermes Agent","1.0.0",[33,34,35,36,37,38,39],"writing a song","song lyrics","music prompt","suno prompt","parody song","adapting a song","AI music generation",[41,42,43],"linux","macos","windows","main","94a12cab2ce32e2020d179b5d579d0cbcbd0f206","hermes\u002Fskills\u002Fcreative\u002Fsongwriting-and-ai-music","https:\u002F\u002Fgithub.com\u002Ffengwhang\u002Fmercury","https:\u002F\u002Fgithub.com\u002Ffengwhang","2026-10-09 08:26:42",{"source":8,"skillName":5,"commands":51,"prompt":61,"zipAvailable":62},[52,55,58],{"tool":53,"command":54},"npx","npx skills add fengwhang\u002Fmercury --skill songwriting-and-ai-music",{"tool":56,"command":57},"bunx","bunx skills add fengwhang\u002Fmercury --skill songwriting-and-ai-music",{"tool":59,"command":60},"pnpm","pnpm dlx skills add fengwhang\u002Fmercury --skill songwriting-and-ai-music","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ffengwhang\u002Fmercury\u002Fblob\u002Fmain\u002Fhermes\u002Fskills\u002Fcreative\u002Fsongwriting-and-ai-music\u002FSKILL.md",true,[],[65,72,79,86,92,99],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":18},"01M7C4BTY3ND38CN200ZCTH6NV","greet","A greeting skill for testing","https:\u002F\u002Fgithub.com\u002Ffengwhang\u002Fmercury\u002Fblob\u002Fmain\u002Fomp\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Fmarketplace\u002Ffixtures\u002Fvalid-marketplace\u002Fplugins\u002Fhello-plugin\u002Fskills\u002Fgreet\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"01M7C4BSNZTPW8JR5NGCQXGSEZ","valid-skill","A valid skill for testing purposes.","https:\u002F\u002Fgithub.com\u002Ffengwhang\u002Fmercury\u002Fblob\u002Fmain\u002Fomp\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Fvalid-skill\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"01M7C4BRA7J8C06KSVW100F1W1","unknown-field","A skill with an unknown frontmatter field.","https:\u002F\u002Fgithub.com\u002Ffengwhang\u002Fmercury\u002Fblob\u002Fmain\u002Fomp\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Funknown-field\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":18},"01M7C4BNB04KQGCJ835JE8AR13","no-frontmatter","https:\u002F\u002Fgithub.com\u002Ffengwhang\u002Fmercury\u002Fblob\u002Fmain\u002Fomp\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Fno-frontmatter\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":18},"01M7C4BM38TZFYT1AJ6SKHPZZ2","child-skill","A nested skill in a subdirectory.","https:\u002F\u002Fgithub.com\u002Ffengwhang\u002Fmercury\u002Fblob\u002Fmain\u002Fomp\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Fnested\u002Fchild-skill\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":18},"01M7C4BJDN7RFFCM28CGX15Q97","different-name","A skill with a name that doesn't match the directory.","https:\u002F\u002Fgithub.com\u002Ffengwhang\u002Fmercury\u002Fblob\u002Fmain\u002Fomp\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Fname-mismatch\u002FSKILL.md",[],[]]