Menu
Skills in this repository
SkillsMP has collected 150 skills from CleanExpo/skills-library. Open a skill to review its source and details.
CleanExpo/skills-libraryShowing 40 of 150 collected skills.
QA Lead — holds every deliverable against a pass/fail rubric before it ships to a client or merges to main. Accepts code PRs, content pieces, SEO reports, and design outputs. Returns PASS or FAIL with specific, actionable reasons. Updates the board_mandates…
Capability-tiered autonomy gating for all agent action — maps the DeepMind AGI→ASI continuum (arXiv:2606.12683) onto four autonomy tiers (L0 advise → L3 strategic/irreversible), each with an explicit gate. Use to decide "can this agent do this action…
Use whenever the Claude-in-Chrome extension reports "Browser extension is not connected" or list_connected_browsers returns empty, before retrying browser work — and when setting up a new machine's Chrome for Claude. Deterministically aligns Chrome's…
Prevents the three recurring security failures distilled from portfolio incidents across dr-nrpg, synthex, ccw-crm, and Pi-Dev-Ops: (1) secrets exposed in tracked docs/runbooks/scripts, (2) environment variable misreads in Vercel Next.js API routes causing…
Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells,…
Use when orchestrating multi-round agent work — a second task in a domain an agent already served, any iterate-review loop (UI rounds, fix cycles), or before spawning a fresh sub-agent for follow-up work. Encodes the warm-specialist pattern - resume via…
Chart a foggy, bigger-than-one-session effort as a shared map of decision tickets on Linear, then work the map one decision per session until the way is clear.
Query the 1000 Genomes Project dataset (3,202 whole-genome-sequenced individuals, GRCh38) at the level of individual participants. Use when a question is about individuals or variants in the 1000 Genomes Project cohort: which individuals carry variants…
Use Parallel CLI for web search, URL extraction, deep research, structured data enrichment, entity discovery, and recurring web monitoring. Best for requests that explicitly need current web evidence, academic-source discovery, repeated entity lookups,…
Compile current scholarly evidence for a scientific manuscript or research brief. Use when the user explicitly asks to gather literature, references, background evidence, competing findings, or a manuscript research packet. Uses Parallel Search by default,…
Zero-shot time series forecasting with Google's TimesFM foundation model. Use for any univariate time series (sales, sensors, energy, vitals, weather) without training a custom model. Supports CSV/DataFrame/array inputs with point forecasts and prediction…
Entry point for 149 vendored scientific research skills (K-Dense Scientific Agent Skills). Use for ANY task in bioinformatics/genomics (scanpy, biopython, pysam, phylogenetics, RNA-seq, variant calling), cheminformatics/drug discovery (rdkit, deepchem,…
Design or review any skill to the Library standard — frontmatter, structure, steering, and pruning.
Use BEFORE designing or building ANY UI — page, screen, component, dashboard, or app flow (founder directive 2026-07-13 — every UI build starts from the Mobbin library) — and when the user mentions Mobbin, UI/UX reference, design inspiration, or asks how real…
Operator command to drill ONE topic to primary sources with current data — iterative multi-agent research that deepens each round until the topic saturates, delivered as a founder-legible decision over a fully cited report; watch mode re-runs on an interval…
Use when a project needs Apify-powered web, social, SERP, marketplace, or competitor research with credential, compliance, and dataset controls.
Build a rights-aware industry image REFERENCE library (research only — never republish) to design ORIGINAL images from. Query-expanded discovery via Exa/Firecrawl/Apify with browser fallback, canonical+perceptual dedup, licence classification + quarantine,…
Generate an on-brand, schema-marked website section from a Google Business Profile location, gated by the estate's copy validators. Dispatched by ai-website; not a standalone entry point.
Build an "AI website" for a local business — a normal-looking site with an AI agent, lead capture, follow-up, and a CRM under the hood. Entry-point orchestrator for the AI-Websites product line. Triggers on "AI website", "build an AI website", "GBP to site",…
Turns one idea into a hook-first, 9:16 short-form video and repurposes it into 8 platform-native cuts — YouTube (Shorts) and Instagram (Reels) first, then TikTok and the rest — grounded in live trend + competitor intelligence. Use when a brief mentions…
The estate's wiki-first recall gate. Before answering, researching, or planning ANY task, deterministically pull grounding from the 2nd-Brain wiki via brain.js — score every estate index (vault OKF + MEMORY.md + skills index) without opening files, open the…
Use when a new skill must be created from a decision, verdict, or capability gap — the user says "forge a skill", "turn this verdict into a skill", "build a skill for X", or pantheon/nexus hands over a VERDICT recommending one. Produces a secure, evaluated,…
Use when a genuinely contested, high-stakes, or novel question deserves a council of the greatest relevant minds — the user says "/pantheon", "convene the council on X", "what would the greats say", or nexus routes a decision that one specialist cannot carry.…
Aligns Margot's knowledge corpus with the latest 2nd Brain wiki, Pi-CEO health data, and empire context. Runs sync scripts, verifies Supabase tables are current, and sends Margot a context refresh prompt. Use weekly or after major wiki updates. Model:…
Use when a project, repo, or knowledge bot (CARSI, RestoreAssist, CCW-ERP, MARGOT, any client brain) needs its knowledge tree brought under the Second-Brain Standard — the user says "adopt the standard", "index this knowledge tree / docs", "put X under the…
Invoke before reporting a task "done", "complete", "fixed", "green", or "shipped" — the final gate that blocks a completion claim unless a this-session tool result proves the desired state was reached, tested, and left green. Prevents CASHE (incomplete /…
Transform uncertainty into decisions through structured board deliberation. Nine specialist personas — CEO, Revenue, Product Strategist, Technical Architect, Contrarian, Compounder, Custom Oracle, Market Strategist, and Moonshot — debate a brief and the CEO…
B.U.I.L.D. Inflow orchestration — run the data-capture pipeline as one routine (mine session lake → refresh OKF indexes → sync wiki to Supabase). Use when the user says "/data-ingestion", "run the ingestion routine", "pull in new data", or on the Tue/Fri…
Generates an 8-week Google Business Profile posting calendar — 2 GBP posts per week, varied across Offers / Events / Updates / Products, each pre-written with a CTA + image brief + UTM-tracked link. The recurring retention asset that compounds local rankings…
The 15+ year brand director extension of `brand-guardian` specifically for video — frame-by-frame audit of brand-mark integrity, Gun Metal `#1a1a1a` + Candy Red `#dc143c` consistency, NO AI slop, NO Lucide icons, custom geometric marks ONLY, real logos…
The 15+ year colorist persona (Light Iron / Company 3 / The Mill-tier) that applies per-brand LUT + grade pass to every Unite-Group video. Closes the "raw Remotion render colours" gap — today Remotion outputs whatever colours BrandConfig declares with zero…
The 15+ year Creative Director persona that owns brief intake, composition selection, and narrative POV for every Unite-Group video. ENTRY POINT for the Video Agency — replaces `remotion-orchestrator` as the brief receiver. Use when the user says "make a…
The 15+ year editor persona (Walter Murch-tier — the Apocalypse Now / The Conversation editor; "In the Blink of an Eye" framework) that owns assembly, cut points, b-roll-to-talking-head ratio, pacing rhythm, J-cuts + L-cuts, and audio sync polish. Closes the…
The 15+ year sound designer persona (Skywalker Sound / Sound Lounge / Audio Network-tier) that mixes ElevenLabs voiceover with music bed, sound effects, ambient texture, and audio polish. Closes the single biggest gap in the existing Remotion pipeline — today…
E-E-A-T specialist — a first-class UPSTREAM consult lens that scores content for Google's Experience, Expertise, Authoritativeness, and Trust signals BEFORE it's finalized, and returns structured fixes. Use when producing or auditing any content, page, or…
Mandatory pre-build challenge gate (/judge). Run before approving or building any feature, connector, automation, agent, hook, MCP server, UI change, database change, or architecture plan. Read-only — performs first-source evidence review, devil's advocate…
Capture real, credible data from authoritative sites into the 2nd-Brain vault Sources/ library so it can be cited by any FACT/non-fiction content (faceless-video FACT mode, marketing copy, SEO articles, client deliverables). Tiered research substrate (margot…
Audit and reclaim a session's context/token budget in one pass — read /context, trim MCP tools, clear vs compact, CLI-over-MCP, model ladder, red-MCP triage.
Resume work from a session handoff. Reads the latest handoff, verifies current repo state against it, reconciles any drift, then continues the work from the documented pickup point without re-deriving old context. Verification is read-only and mandatory…
B.U.I.L.D. Loop — the improvement engine. Scans the 2nd Brain vault + session digests for fixable problems (broken wikilinks, orphan sources, near-duplicate pages, session signal) and emits GRADED proposals into 3 buckets (auto-apply / need-signoff /…