Velith / rules
Cursor rule40 starsChanged 2 months agoApache-2.0View on GitHub
Velith genre-specific writing conventions for each supported genre
In this repository23 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- dashboard/CLAUDE.mdCLAUDE.md
- .cursor/rules/velith-editing.mdcCursor rule
- .cursor/rules/velith-pipeline.mdcCursor rule
- book-academicSkill
- book-draftSkill
- book-editSkill
- book-fictionSkill
- book-gameSkill
- book-genre-creatorSkill
- book-ideationSkill
- book-illustrateSkill
- book-initSkill
- book-nonfictionSkill
- book-outlineSkill
- book-poetrySkill
- book-publishSkill
- book-screenplaySkill
- book-statusSkill
- book-technicalSkill
- book-visualsSkill
- loomSkill
The file
Files
--- description: Velith genre-specific writing conventions for each supported genre globs: - "**/drafts/**" - "**/outline.md" - "**/PRD.md" alwaysApply: false --- # Velith — Genre Conventions ## Fiction - Plot structure: Save the Cat! 15-beat or Snowflake Method - Character bible with GMC (Goal, Motivation, Conflict) per character - Scene beats with RDD (Reaction, Dilemma, Decision) pacing - POV consistency (no 1st/3rd mixing within chapters) - Dialogue rules: subtext over exposition, distinctive voice per character ## Non-Fiction - Problem-solution structure per chapter - Learner persona–driven progression (what does the reader already know?) - Evidence hierarchy: primary research > expert interviews > case studies > anecdotes - Chapter hooks: open with a question or counterintuitive fact - Summary + action items at chapter end ## Technical - Concept progression gradient: novice → intermediate → expert - Code examples: compilable, commented, with expected output - Diagrams for architecture, data flow, or process steps - API docs: endpoint, parameters, response schema, error codes - Labs/exercises at end of each chapter ## Screenplay - 3-act structure + 8-sequence method - Dialogue: subtext, distinct voice, no on-the-nose exposition - A/B story interweaving (main plot + subplot) - Action lines: present tense, visual, concise - Page budget: ~1 page = ~1 minute of screen time ## Poetry - Form types: sonnet, haiku, free verse, villanelle, prose poem, etc. - Meter/rhythm awareness (iambic, trochaic, anapestic, dactylic) - Imagery systems: visual, auditory, tactile, olfactory, gustatory - Collection arc: ordering poems for emotional/thematic progression - Line break intentionality: enjambment vs end-stopped ## Game Scenario - Quest/event tree structure with branching conditions - Dialogue branching with player agency flags - Lore bible: world rules, factions, technology/magic systems - Flag system: track player choices across chapters/scenes - Multiple endings: foreshadowing and divergence points ## Academic - IMRAD structure (Introduction, Methods, Results, Discussion) - Literature review: thematic organization, not chronological - Argument chains: claim → evidence → warrant → backing → rebuttal - Citation practices: consistent style (APA, MLA, Chicago, etc.) - Abstract and keywords before main body ## Custom Genres - Compose patterns from any supported genre via genre-creator - Define custom quality gates and chapter templates - Can inherit from multiple parent genres ## Visuals by genre Every genre skill carries a Visuals section. Fiction: optional, character sheets locked before scene art, headers as motifs. Non-fiction: figures are part of the argument, charts from data files, photos licensed. Technical: diagrams as code with one theme, real screenshots, ≤3 figures per chapter. Screenplay: pitch packages only. Poetry: restraint, one medium, spot art at section openings. Game: maps, branching flowcharts from quest data, locked character and location sheets. Academic: data figures from `sources/`, discipline conventions. All through `/book-visuals`; the art bible is the single source of look.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

