Skill library
Publisher description: The Claude Code operating stack for AI-startup go-to-market: skills, slash commands, subagents, deterministic pipelines and an MCP server sharing one context layer. Competitor intel, content production, SEO backlog, lead-finding, paid-media craft.
Catalogue scope: 6 procedural entries from 6 source paths.
Repository health
GitHub’s own figures, read from its REST API. Commit and contributor totals come from the paging header rather than an estimate.
Repository metadata checked 2026-10-12 · skill files extracted 2026-10-06 · published 2026-09-05
Extraction and automated analysis describe the files at the recorded revision. They are not an editorial review or a test of the skills in use.
- Stars
- 0
- Commits
- 4
- Last 90 days
- 4commits
- Contributors
- 1
- Last push
- 2026-09-22
- First commit
- 2026-09-05repo created
6 skill entries
- brand-positioning
Define a brand's strategic foundation — the internal identity that drives every external expression. Produces Brand Purpose, Values, Personality, Target Audience, Brand Promise, and a classic Positioning Statement, written to docs/brand-positioning.md. Can also run named positioning frameworks on demand — April Dunford (Obviously Awesome), Geoffrey Moore's statement, Ries & Trout, Jobs-to-Be-Done, Category Design (Play Bigger), and the Value Proposition Canvas. Use when the user wants to define, refine, or pressure-test WHO a brand is (its identity), build positioning from scratch, extract it from existing materials, reposition, or apply a specific positioning framework by name. This is the foundation tool that competitive-positioning, brand-voice, and ux-brief build on. NOT for competitor comparison (that's competitive-positioning) or voice/tone guides (that's brand-voice).
Self-containedInstructions only2,444 words - find-blog-opportunities
Discover and score keyword + topic opportunities for Notch SEO blogs, grounded in the product and existing keyword clusters. Outputs a prioritized backlog (CSV + markdown) that feeds Step 1 of the blog pipeline. Use when the user wants blog topic ideas, keyword opportunities, SEO content gaps, "what should I write about", a content backlog, or to find keywords to rank for in Notch's space. NOT for writing/publishing a blog (that's your blog pipeline) or for ad-hoc keyword lookups.
Needs a keyActs undeclared2,557 words - format-radar
Run one episode of the recurring "Max Format Radar" Twitter/X series end to end: research a trending ad format, deconstruct why it converts, decide which category it's already burned in, mutate it into a Max-ownable version, generate the scenes on Higgsfield, build the 1080x1920 multi-scene post video, and write the 5-post thread. Use when the user drops a new ad format ("do a format radar on X", "new format: street polls", "make a post about this ad format"), names a category to go find one in, or asks for the next episode of the series. NOT for general blog work (see your blog pipeline) or one-off Higgsfield generations (see higgsfield-generate).
Needs a keyActs undeclared4,663 words - launch-post
Write a Notch launch/feature post for X (Twitter) from a video script, product concept, or founder draft — then produce the matching creator amplification brief, quote packs, and comment packs. Encodes the locked Notch post structure (3-line hook: specificity + open curiosity → widen → take), the hook-mechanics library, the iteration protocol (variants → ranked table → pick), the hard voice rules (no invented numbers, no em dashes, hire language for Max), and the creator-brief format (vault MD + Notch-yellow artifact page + guardrails). Use when the user says "create a post around this script", "write the launch post", "give me hook variations", "make a creator brief", "post for the video going out tonight", or pastes a draft post for tightening. NOT for SEO blogs (your blog pipeline), Max Format Radar episodes (format-radar), or founder tweet replies (/founder-voice-reply).
Self-containedActs undeclared1,600 words - product-grading-report
Recurring niche workflow grading report that tells Notch what to build next. Input: a niche (e.g. "media buying"). Maps the ideal end-to-end workflow, grades every step on demand / tool coverage / DIY pain / trajectory / Notch fit, prices the paid stack vs the free-DIY path, places Notch on the map, and outputs a Build-next / Defend / Don't-build / Watch verdict plus a diff vs the last run. Use when the user says "grade the market for X", "product grading report", "where does Notch stand in X", "what should we build next", "rerun the media buying report". NOT for SEO blogs (find-blog-opportunities), positioning copy (notch-positioning-brainstorm), or single-tool teardowns.
Self-containedActs undeclared774 words - x-content-playbook
Performance-grounded rules for writing and planning posts on the founder's X account — hook selection, CTA decision, angle rotation, copy skeleton, cadence — derived from the measured Jul–Aug 2026 post population (11 posts, views/comments/saves per post). Use when writing an X post or thread for the account, planning a content calendar week, choosing between concepts, or reviewing a draft before it ships ("write the post", "which concept should run", "review this draft", "plan next week's posts"). NOT for SEO blogs (your blog pipeline), Format Radar episode production (format-radar), or launch-post mechanics (launch-post — that skill owns the locked launch structure; this one owns the performance rules both must obey).
Self-containedInstructions only1,363 words
Need help setting it up?
This page tells you what Ajitesh-png/ai-gtm-stack does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.
Book a call →The directory stays free. There is nothing gated behind this.