fltman

Synthesize research reports into 3-5 concrete, ranked artist concepts for the AI artist factory. Use this skill after all research skills have produced their reports (Phase 2), when you need to generate artist concepts from data, or when the manager requests new concept ideas. Every idea must be grounded in research data, not personal taste.

Turn a chosen artist concept into a complete, production-ready concept bible at docs/concept-bible.md. Use this skill after the brainstorm skill has produced concepts and one has been selected (Phase 2 completion), or when the manager needs an updated concept bible. The bible must be detailed enough that any production skill can create on-brand content without additional guidance.

Generate AI images using Google Gemini via OpenRouter. Use when the user wants to create images, generate pictures, make AI art, edit images, or transform photos. Supports both text-to-image and image-to-image generation.

Write all text content for the AI artist — social captions, bios, press releases, cryptic messages, interviews, fan responses. Use this skill when the manager delegates writing, when creating captions for posts, when drafting press releases, or when crafting the artist's voice in any text form. Always read docs/concept-bible.md first for voice and tone.

Create all visual assets for the AI artist — artist photos, album covers, social content, event documentation, studio glimpses. Use this skill when the manager delegates visual work, when creating artwork for releases or social posts, or when generating event photos for the tour hack. Requires the imagegen skill. Always read docs/concept-bible.md first for visual identity.

Orchestrate all production skills — plan sprints, delegate tasks, track status, review outputs. Use this skill when starting Phase 4, when asking "what should we do next", when planning a release or event campaign, or when any coordination between production skills is needed. This is the primary entry point for all production work.

Plan release cycles, growth strategy, playlist/blog targeting, event campaigns, and brand trajectory for the AI artist. Use this skill when planning a release rollout, targeting playlists or blogs, designing an event campaign, setting growth milestones, or when the manager requests strategic planning. Always read docs/concept-bible.md first.

Generate Suno-ready song packages (300-character style description + tagged lyrics) for the AI artist. Use this skill when creating new tracks, when the manager delegates music production, or when anyone asks to generate music for the artist. Requires the Suno music skill to be installed. Always read docs/concept-bible.md first for sound profile.

Create specialized AI agents for Claude Code projects. Use when the user wants to build a new agent, create a team of agents, add an agent to .claude/agents/, or needs help designing agent roles for a project.

Create voice style descriptions for ElevenLabs. Use when the user wants to describe a voice, create a voice profile, or generate voice style text. Outputs exactly 300 characters describing voice characteristics.

Generate AI images using Google Gemini via OpenRouter. Use when the user wants to create images, generate pictures, make AI art, edit images, or transform photos. Supports both text-to-image and image-to-image generation.

Generates AI videos from text prompts or reference images using OpenAI Sora. Use when the user wants to create videos, generate video clips, make AI films, animate images, or produce motion content. Supports text-to-video, image-to-video, and video remixing.

Create songs for Suno AI music generation. Use when the user wants to create a song, write lyrics, or make music. Outputs a 300-character style description and tagged lyrics formatted for Suno.

Set up new Claude Code projects with CLAUDE.md, slash commands, and proper structure

Set up, scaffold, and administer agentic loops in Claude Code — headless while-loops around `claude -p`, evaluator-optimizer (generator/critic) loops, meta / prompt-refinement loops (a loop that refines the prompts another loop runs), and orchestrator fan-out. Every loop ships with non-negotiable guardrails: a verifiable exit condition, a max-iteration cap, and in-code budget enforcement. Use whenever the user wants to build, design, run, wire, or supervise loops / agent loops / "loop engineering" / nested sub-loops, or asks how to make Claude prompt Claude on a schedule or until a goal is met.

Generate natural voice-over / narration audio with the ElevenLabs Text-to-Dialogue API. Use when producing a narrated demo video, podcast, audiobook, or any per-line voice-over where each line becomes its own mp3 and identical lines are cached (no re-billing). Sounds markedly better than plain text-to-speech. Reads ELEVENLABS_API_KEY from the environment.

Teknisk referens för PDF-till-agent transformation

Nybörjarvänlig GitHub-hjälp som förklarar allt utan teknisk jargong. TRIGGA DENNA SKILL när: - Användaren nämner git, github, commit, push, pull, branch, merge - Användaren verkar osäker på versionshantering - Användaren frågar "hur sparar jag", "hur delar jag", "hur backar jag" - Användaren vill "starta ett projekt", "bygga något", "göra en app" - Felmeddelanden innehåller "git", "repository", "commit", "merge conflict" - Användaren är nybörjare och arbetar med kod Skillen ger: Liknelser, enkla förklaringar, proaktiva förslag, och steg-för-steg guider på svenska utan tekniskt språk.