seldonframe

In the author's own words, from each skill's SKILL.md.

• block-creation

  Scaffold a new SeldonFrame block from a natural-language intent. Use when the builder says "build me a block that ...", "scaffold a block that ...", or similar. Produces a real, validated block skeleton in packages/crm/src/blocks/ — BLOCK.md with composition contract, tools.ts with Zod schemas, subscription handler stubs (when reactive), and test.todo stubs per tool. Does NOT commit, does NOT ins…

  skills/block-creation/SKILL.md on GitHub (opens in a new tab)

• landing-page-creation

  Build a SeldonFrame workspace's marketing landing page using the existing block primitives (get_block_skill + persist_block). Use when the operator asks to "build a landing page", "make a website", "design the home page", "redo the landing", or similar. Triggers AFTER workspace has been created via create_workspace_from_url (which by default creates only a chatbot- preview demo page — this skill…

  skills/landing-page-creation/SKILL.md on GitHub (opens in a new tab)

• seldonframe-agent-business

  Use when a builder working in an IDE agent (Claude Code, Cursor, Codex) wants to build, test, deploy, sell, or get paid for an AI agent on SeldonFrame — e.g. "build me a receptionist agent and sell it", "deploy my agent to a real phone number", "list my agent on the marketplace", "run my agent's evals", "withdraw my agent earnings" — or when the SeldonFrame MCP (@seldonframe/mcp) is connected and…

  skills/seldonframe-agent-business/SKILL.md on GitHub (opens in a new tab)

• seldonframe

  No description in the SKILL.md.

  skills/seldonframe/SKILL.md on GitHub (opens in a new tab)

• deploy-prod

  Deploy main to Vercel production the safe way (clean worktree, env baked in) and smoke it with Reelier at Level 0 — zero tokens. Use after any merge to main; Vercel git integration does NOT reliably auto-deploy main.

  .claude/skills/deploy-prod/SKILL.md on GitHub (opens in a new tab)

• dream

  Daily out-of-band reflection — mine the last 24h of SeldonChat vision_check failures, cluster them, and PROPOSE (never auto-apply) lessons / tool-desc / schema-gap fixes, with a measured pass-rate delta. Run once/day via a schedule, or manually to reflect now. The Telescope→self-improvement loop, scoped to what we can verify honestly.

  .claude/skills/dream/SKILL.md on GitHub (opens in a new tab)

• extract-approach

  The recorder — after every non-trivial solved problem, write ONE learnings note capturing the approach, the judgment calls, and the reusable rule, written for a weaker model reading cold. A solution without its learnings note is unfinished work.

  .claude/skills/extract-approach/SKILL.md on GitHub (opens in a new tab)

• generate-and-grade

  Run the agent-generator regression net. Use after ANY change under packages/crm/src/lib/agents/generate/** (parse-intent, agent-bundle, bind-tools, tool-catalog) to catch generator breakages — misclassification, lost/wrong tool bindings, wrong channel, wrong skill template — before a human does.

  .claude/skills/generate-and-grade/SKILL.md on GitHub (opens in a new tab)

• handoff

  Write a session handoff that lets a fresh agent (or you after compaction) resume cold with zero re-discovery. Use when ending or pausing a work session, before a likely compaction, or when passing work to another agent/worktree. Captures STATE (what is true now) over INSTRUCTIONS (what to do) — state survives when plans change; step-lists rot.

  .claude/skills/handoff/SKILL.md on GitHub (opens in a new tab)

• information-gain

  The non-commodity sourcing step for any SEO/content loop — find niche YouTube videos, pull their transcripts, and mine ORIGINAL founder stories / real numbers / failures into cited source briefs that exist nowhere else in writing. Google's helpful-content system ranks information gain; this is how a loop produces it instead of commodity AI text. Invoke as a sub-step from x-post-engine, content-lo…

  .claude/skills/information-gain/SKILL.md on GitHub (opens in a new tab)

• reflect

  Think through any decision — code, business, personal — and get a straight answer. Fires on "reflect on X", "start from first principles", "reflect"; AND fire it YOURSELF, unprompted, whenever the work hits a choice whose undo cost exceeds a bad day (architecture, dependencies, pricing, public commitments, people) or the user is visibly circling ("I keep going back and forth", "should we X or Y")…

  .claude/skills/reflect/SKILL.md on GitHub (opens in a new tab)

• ship-feature

  Run the full SeldonFrame feature loop — brainstorm → spec → plan → subagent build → verify-build → merge → memory — stopping only at the two human gates (spec approval + merge). Invoke with a one-sentence feature description.

  .claude/skills/ship-feature/SKILL.md on GitHub (opens in a new tab)

• verify-build

  SeldonFrame merge gate — run unit tests + tsc + check-use-server + migration-journal + regression-grep and return ONE pass/fail verdict. Invoke before merging any branch; nothing merges without a green verdict.

  .claude/skills/verify-build/SKILL.md on GitHub (opens in a new tab)

• vision-verify

  Visually verify a rendered web page against a goal + rubric — screenshot it (reliable off-box render, NOT the flaky Chrome-MCP), then grade the pixels with an INDEPENDENT vision pass. Use for any UI/visual change before calling it done, or to confirm a fix landed on a live/preview page. Catches what verify-build + code review can't SEE: contrast, overlap, broken images, layout shift, duplicated e…

  .claude/skills/vision-verify/SKILL.md on GitHub (opens in a new tab)

• x-post-engine

  Mine the current work session (or a named event) into 2-3 draft X posts using the 6 founder-content formats, keyword-first. Drafts only — Max makes the angle call and posts. Invoke at the end of a build session or with a topic.

  .claude/skills/x-post-engine/SKILL.md on GitHub (opens in a new tab)

• youtube-video-kit

  Produce the complete publishing kit for one keyword-targeted YouTube video (script, thumbnail, title/description/chapters/pinned comment) from its matching SEO page, and wire the videoId back into the page after publish. Invoke with a /best or /tools slug, or a keyword.

  .claude/skills/youtube-video-kit/SKILL.md on GitHub (opens in a new tab)

• website-chatbot

  Friendly, professional chat assistant for service businesses (HVAC, dental, coaching, agency, accounting, etc.). Answers FAQ from operator-curated knowledge, books appointments via the same booking primitive that powers /book, escalates to human via portal-message when out of scope.

  packages/crm/src/agents/website-chatbot/SKILL.md on GitHub (opens in a new tab)

• about

  About-the-business section — who they are, why they started, what makes them specific. Trust-building, not corporate.

  packages/crm/src/blocks/about/SKILL.md on GitHub (opens in a new tab)

• booking

  The booking page (calendar) — what visitors see when they hit /book. Title, description, slot duration, location kind, weekly availability hours, and the form fields collected at booking time.

  packages/crm/src/blocks/booking/SKILL.md on GitHub (opens in a new tab)

• cinematic-landing

  Dark cinematic landing system — looping Pexels MP4 background + liquid-glass UI + Instrument Serif italic typography + Framer Motion choreography. Auto-applied to agency + cinematic-aspirational (coaching, medspa, wellness, luxury) archetypes during create_full_workspace. The marketplace-forkable "fat skill" that captures the entire Aura / Velorah / Aethera / Asme aesthetic family.

  packages/crm/src/blocks/cinematic-landing/SKILL.md on GitHub (opens in a new tab)

• composite

  Universal landing-page section built from low-level primitives. Use for any block that doesn't fit hero/services/about/faq/mid-cta — comparisons, pricing tiers, "how it works," stats rows, side-by-side image+text, custom CTAs, "vs the competition."

  packages/crm/src/blocks/composite/SKILL.md on GitHub (opens in a new tab)

• cta

  Mid-page call-to-action — a focused conversion moment between sections. Singular outcome, low friction, urgency without pressure.

  packages/crm/src/blocks/cta/SKILL.md on GitHub (opens in a new tab)

• faq

  Frequently-asked-questions section addressing the actual concerns that block visitors from booking — pricing, scheduling, what to expect, scope, refunds.

  packages/crm/src/blocks/faq/SKILL.md on GitHub (opens in a new tab)

• hero

  Above-the-fold hero with quantified value claim, primary CTA, and supporting visual.

  packages/crm/src/blocks/hero/SKILL.md on GitHub (opens in a new tab)

• intake

  The intake form (the lead-capture page at /intake). Title, description, and the questions visitors answer. Per-vertical — a barbershop intake is short, a legal intake is structured, a coaching intake is exploratory.

  packages/crm/src/blocks/intake/SKILL.md on GitHub (opens in a new tab)

• services

  Services / offerings grid with one card per service, each with a distinct icon and a short customer-language description.

  packages/crm/src/blocks/services/SKILL.md on GitHub (opens in a new tab)

No skill matches that filter.

Similar skills

• ponytailby DietrichGebert Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote. 160k stars · 6 skills
• archifyby tt-a1i Turn any idea, plan, or codebase into a beautiful interactive diagram. An agent skill for Claude Code, Codex, and more. 81.2k stars · 2 skills
• awesome-claude-skillsby ComposioHQ A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows 76.8k stars
• marketingskillsby coreyhaines31 Marketing skills for Claude Code and AI agents. CRO, copywriting, SEO, analytics, and growth engineering. 53.9k stars · 30 skills
• reaby morluto Reverse engineer anything with agents, from app behavior down to native binaries. 46.4k stars · 1 skill
• awesome-agent-skillsby VoltAgent A curated collection of 1000+ agent skills from official dev teams and the community, compatible with Claude Code, Codex, Gemini CLI, Curso… 35.4k stars