AI Skills Library · plain Markdown · load them anywhere
Every file here is a working method, not a plugin — what the task is, what it is not, the step-by-step method, a quality bar, and guardrails that route regulated judgment to licensed professionals. Because it is plain text, a skill loads into Claude, any chat assistant, or a local open-weight model with zero integration. Two kits: a consumer kit bound to the Consumer Design System's machine contracts and an institutional kit from the regulated-finance design system.
Claude
ChatGPT
Codex
Gemini
Antigravity
Cursor
Figma MCP
Grok
DeepSeek
Qwen
Hermes Agent
OpenClaw
NVIDIA
Hugging Face
Ollama
n8n
Dify
Cline
OpenCode
Z.ai
Eigent AI
Copilot
AutoGPT
Mem0
Voyager
Manus
Pydantic AI
CrewAI
OpenHands
Claude
Cowork
ZeroClawMachine-readable index for agents: cds/components.json (skills enumerated with URLs) · drop-in brief: cds/AGENTS.md · compile a custom pack with the AI-Ready Export. Skills are methods with honest guardrails — not licensed legal, tax, securities, medical or clinical advice.
Turn a company's scattered history — tickets, epics, reports, chat, version history, UAT and release records — into a cited, role-scoped, continuously maintained institutional memory. Decision archaeology with provenance, so domain knowledge survives attrition and new people ramp in days, not quarters.
Download .md ↓Verify a consumer surface against the accessibility contracts every CDS entry carries — WCAG 2.2 AA as the floor: target sizes, reduced motion, live regions, focus order, accessible authentication.
Download .md ↓Design and build AI features with the 22-entry AI & Agents domain — gates that sign exact payloads, traces that record blocks, streaming that stays accessible, honest capability disclosure.
Download .md ↓Add a new component to the Consumer Design System the way the existing 142 were built — four-cell register first, tokens only, delegated events, reduced-motion contract, and a platform anchor it answers to.
Download .md ↓Audit a consumer flow against the system's hard rules and the public rulebooks behind them — consent symmetry, honest pricing, click-to-cancel, review integrity, engagement pressure — with findings a regulator would recognise.
Download .md ↓Review a mockup, prototype, or PR against the Consumer Design System's registers — a structured pass that cites entries instead of taste, for design crits and code review alike.
Download .md ↓Port CDS components to React, SwiftUI, Compose, or Vue while keeping the contract — the register travels, the tokens map, the accessibility behaviour survives the translation.
Download .md ↓Apply the system's motion contract — one spring, three durations, self-terminating loops, reduced-motion as a contract — including the Text Animations domain's usage limits.
Download .md ↓Analyse any product surface — or specify a new one — into a complete, tokenized Design Kit document: YAML tokens, component specs with token references, do/don'ts, responsive behaviour, and an honest Known Gaps section. The universal method behind a full-site design teardown.
Download .md ↓Compose a complete consumer screen from the Consumer Design System catalogue — selection, bundles, hard rules, tokens, sign-off — for designers and developers working from a product brief.
Download .md ↓Extend, re-brand, or add themes to the token layer without breaking its contracts — light/dark in lock-step, AA on every pairing, semantic names, components never touching raw hex.
Download .md ↓Write microcopy to the Consumer Design System's voice — honest states, the error formula, consent in plain language, buttons that name outcomes — so the words carry the same contract as the components.
Download .md ↓Decomposes a PRD into Epic → Story → Sub-tasks, writes Given-When-Then (BDD) acceptance criteria that are actually testable, and estimates relative story points — so the delivery plan is executable and "done" is checkable rather than arguable. For product manager, scrum master, tech lead.
Download .md ↓Use LLMs and coding agents where they genuinely help — refactoring, debugging, test generation, code understanding — with the prompt patterns and the guardrails that make the output trustworthy: every diff reviewed, tests as the acceptance gate, and a human owning the merge, so AI accelerates engineering without eroding correctness.
Download .md ↓Turns a PRD's business goals into an event-tracking specification — event names, triggers, and parameters — plus the funnel and retention tracking logic behind them, all under a consistent naming convention so the data is analysable instead of a pile of inconsistent events. For product manager, data, growth.
Download .md ↓Analyzes real-time behavioral data, dynamically switches UI layout and CTA variants, tracks statistical significance against a pre-registered hypothesis, and rolls the winning variant out — with guards against p-hacking and peeking, consent-respecting personalization, and a human gate on irreversible or high-risk changes. Results are measured, not assumed. For growth, product, and experimentation teams.
Download .md ↓Design, deploy, and operate AWS infrastructure with an agent — infrastructure-as-code, least-privilege IAM, cost guardrails, and the Well-Architected discipline — so a coding agent can build real cloud systems fast without leaving a security hole or a surprise bill. For developers and platform teams shipping on AWS.
Download .md ↓Reads your product backend analytics — CTR, conversion funnel, drop-off points, retention — to find unconsidered functional gaps and experience breakthroughs, then strengthens both traditional search (SEO) and 2026 generative-engine optimization (GEO) through keyword, semantic-structure, and traffic strategy grounded in your own numbers. For growth, product manager, founder.
Download .md ↓Writes and optimizes blog content with current SEO and generative-engine (GEO) structure, then detects and removes over-narration, empty filler, and machine "AI tone" — tuning the draft to a natural, authoritative, human brand voice without inventing facts. For content, marketing, founder.
Download .md ↓Ship and defend at the edge with an agent — Workers, Pages, R2/D1/KV, caching, WAF, and Zero Trust — with DNS and security changes gated because they take effect globally in seconds. For developers building on Cloudflare's edge and CDN.
Download .md ↓Build a Confluence space people actually find and trust — an information architecture that matches how the team looks for things, page templates that make good docs the default, a single source of truth per topic, and a maintenance rhythm that fights rot. For teams running product, engineering, or ops documentation on Confluence.
Download .md ↓Build analytics and data pipelines on Databricks that stay correct, governed, and cost-controlled — medallion architecture with Delta Lake, jobs and workflows with real failure handling, Unity Catalog governance and PII protection, and cluster/cost discipline. Irreversible data operations gate on a human. For data engineers and analysts working in the lakehouse.
Download .md ↓Get the full value out of Datadog — a coherent observability system across metrics, logs, traces, and monitors, with SLOs, dashboards, and alerting that a person or an agent can operate and manage on the team's behalf, while keeping cost and PII under control. For teams running production systems who want observability that answers questions instead of generating noise.
Download .md ↓Build automated delivery pipelines that ship safely — CI that gates on tests, containerization with Docker, and Kubernetes orchestration — with progressive rollout, rollback, secrets hygiene, and a human on the irreversible, so speed comes from automation and safety from the pipeline, not from luck.
Download .md ↓Help a team building a digital wallet reason through every layer before it touches real value — threat model, key management, encryption and secure storage, transaction integrity, the honest limits of what software can guarantee, and the regulatory surface. A design-and-architecture method with security humility, not a security guarantee. For teams building custodial or self-custody wallets.
Download .md ↓Design cloud-native, distributed systems that hold up — making the hard trade-offs of consistency, partitioning, and failure explicit, choosing serverless vs containers vs cluster by the actual load, and automating infrastructure — so the architecture matches the requirement instead of chasing scale it doesn't have.
Download .md ↓Serve one open-source base model as many specialists — an intent classifier hot-swaps task-specific LoRA adapters (PRD authoring, code refactoring, HR, and more) at inference time, with cached adapters, a confidence threshold, and a base-model fallback, so a single local model gets specialized depth per task without misrouting or over-claiming general expertise.
Download .md ↓Build and maintain a Firebase backend with an agent — security rules first, Emulator-tested, App Check-protected, and cost-aware — so a Flutter or web app gets auth, data, functions, and storage that are fast to ship and safe to run. For developers maintaining a Firebase server layer.
Download .md ↓Raise software quality with the rigorous end of the toolbox — static analysis, property-based and contract testing, and lightweight formal specification where correctness really matters — plus a QA strategy that puts effort where the risk is, so critical logic is verified rather than just spot-checked.
Download .md ↓Reviews GitHub trending and relevant repositories against a team's current needs and bottlenecks, recommends existing open-source projects instead of reinventing the wheel, and — for every recommendation — checks the open-source license (MIT / Apache-2.0 versus strongly-copyleft GPLv3) and flags the risk for a commercial closed-source product. For engineering lead, CTO, developer.
Download .md ↓Run a repository like an engineering org with an agent — PR-first flow, CI as the merge gate, branch protection, CODEOWNERS, Dependabot, and release discipline — so an agent can move fast in the repo without anything unreviewed reaching main. For developers orchestrating GitHub with AI.
Download .md ↓Set up Google Analytics 4 to answer real product questions instead of collecting vanity metrics — a measurement plan tied to decisions, correct event and conversion modelling, consent-aware and privacy-compliant collection, and honest reading of the data. For founders, growth, and product teams instrumenting a site or app.
Download .md ↓Delegate work to an asynchronous coding agent like Google Jules and get back reviewable PRs — crisp task specs, a plan you approve before it runs, tests as the acceptance gate, and a QA-automation plan that turns edge cases into a standing suite. For developers running async agents and building test coverage.
Download .md ↓Choose, license-check, and deploy models from the Hugging Face Hub responsibly — read the model card before the weights, verify the licence before commercial use, and pick the right serving path (Transformers, Inference Endpoints, or TGI) for the load. For developers putting open models into production.
Download .md ↓Run a project in Jira with an agent as a disciplined PM — epics and stories with real acceptance criteria, honest sprint hygiene, JQL that surfaces risk, and status that reflects reality — so planning artefacts stay true instead of becoming a fiction everyone ignores. For anyone using an agent to manage delivery.
Download .md ↓Give an open-source, locally-hosted model real agent execution — ReAct-style prompting plus an API interceptor that lets it trigger Python scripts, database queries, and external webhooks reliably, with schema validation, an allowlist, and a human gate on anything irreversible, so on-prem agents approach cloud-model tool-use without the footguns.
Download .md ↓Use a large-context local model to ingest a legacy project — 100K–300K tokens of source, architecture docs, API specs and migration history — and surface zombie code, dependency-conflict and breaking-upgrade risk, and Clean-Code violations, then propose low-blast-radius refactor PRs that ship with test logic. Runs locally so proprietary code never leaves the machine. For engineers inheriting a system nobody fully remembers.
Download .md ↓Decide whether machine learning or generative AI actually fits a business problem, then frame, evaluate, and deploy it responsibly — with the right metric, honest baselines, and a human on the irreversible decisions — instead of adding a model because it is fashionable.
Download .md ↓Design and operate a MySQL backend that stays correct, fast, and safe — a schema modelled from access patterns, indexes that match real queries, parameterised access that closes injection, least-privilege accounts, and backups proven by restore. Every destructive or schema-changing operation gates on a human and a backup. For developers building on MySQL or MariaDB.
Download .md ↓Build fully-automated end-to-end and integration test pipelines in n8n — scheduled runs, HTTP/API assertions, webhook triggers, contract checks, and fail-loud alerting — so a product's critical paths are verified on a cadence, not by hope. For developers and QA who want tests that run themselves.
Download .md ↓Turn free text — reviews, social posts, support tickets, survey open-ends — into structured signal with natural-language processing: themes, sentiment, and emerging issues, reported with confidence and grounded in real quotes rather than a single averaged mood score.
Download .md ↓Use Notion as a structured thinking-and-writing partner — turn a messy brief into a clear document, decision, or plan by externalising the reasoning into a shape a person and an agent can both work in. Covers workspace and database structure, thinking templates, progressive drafting, and how an agent should collaborate in the doc rather than replace the judgment. For anyone who thinks by writing.
Download .md ↓Deploy Ollama locally with Docker and choose the right local model for the hardware you actually have — a VRAM-first sizing method that reads a GPU's memory budget and recommends the largest model and quantization that will run well, not the one that fits on paper. For developers running on-prem or offline AI.
Download .md ↓Use an answer engine like Perplexity for fast, cited research that speeds discovery without outsourcing judgment — follow the citations to the primary source, separate what's grounded from what's synthesized, and never ship a claim you didn't verify. For anyone using AI search to learn a domain or check a fact.
Download .md ↓Compares a new requirement against your existing architecture, past PRDs, prior R&D records, and product goals to surface logic conflicts, feature overlap, API breaking changes, and database-schema compatibility risk — then writes a rigorous PRD with edge cases and a Jira-ticket breakdown. Reduces re-litigation and late-stage surprises by grounding the spec in what already shipped. For product manager, founder, tech lead.
Download .md ↓Turn a data question into a reproducible Python pipeline — clean, join, transform, automate, and (where permitted) scrape — with pandas/NumPy, notebooks, and scheduled jobs, so the number that ends up in a deck is one you can re-run and defend.
Download .md ↓Turn fuzzy business asks into precise, testable technical requirements and a defensible scope — with explicit boundaries, non-goals, assumptions, and acceptance criteria — so a team builds the right thing once instead of discovering the real requirement in QA or after launch.
Download .md ↓Turn your own site's traffic data into an honest SEO and GEO (generative-engine optimization) improvement plan. Ed Chen's method for auditing a portfolio or product site — fold out your own dev/QA traffic, separate bots from real humans, read timezone and referrer signals, then strengthen both classic search ranking and AI-answer discoverability. For anyone who owns a site and has real analytics to work from.
Download .md ↓Wire Slack as the team's operational nervous system with an agent — specific notifications, incident channels that assemble themselves, in-channel approvals with a human gate, and workflows that reduce noise instead of adding to it — so Slack informs the team rather than training them to mute it. For teams automating ops in chat.
Download .md ↓Choose and run the right statistical model — regression, multivariate methods, hypothesis tests, inference — then report effect sizes and uncertainty honestly, so a decision rests on evidence rather than a p-value taken out of context.
Download .md ↓Turn a Trello board into a project system that actually moves work — a board and list structure that models the real workflow, cards with clear ownership and acceptance criteria, WIP discipline, and automation that removes busywork without hiding status. For small teams and solo builders running product delivery on Trello.
Download .md ↓Turn VS Code into an agent cockpit — tasks and launch configs the agent can run, MCP servers wired to real contracts, reviewed diffs, and workspace-scoped trust — so an AI coding agent inside the editor moves fast against a system it can be refused by. For developers running agents in VS Code.
Download .md ↓Get reliable work out of Cline — the autonomous coding agent inside VS Code with Plan and Act modes and MCP support — by using Plan mode as the checkpoint, approving edits deliberately, and wiring MCP tools like the CDS contract so the agent builds to a real system. For developers working in VS Code.
Download .md ↓Design, build, configure, and finish a Discord bot end to end — command and interaction design, gateway and intents setup, permissions and role model, slash commands, moderation and safety, and deployment. Architected for stability and growth, compliant with Discord's rules. For developers and community builders shipping a Discord bot.
Download .md ↓Use a Hermes-family open-weight model as the reasoning core of an autonomous agent — structured function-calling, tool routing, and a human-gated action loop — so a local, open model can drive real work without becoming an unbounded actor. For developers building agents on open weights.
Download .md ↓Get real work out of Monica — the all-in-one AI assistant that sits across your browser and apps — with a repeatable workflow for research, drafting, translation, and summarization that keeps a human editing the output instead of pasting it. For anyone using a general AI assistant seriously.
Download .md ↓Operate an open-source computer-use / browser-automation agent safely — allowlisted targets, no credentials handed to the agent, human gates on anything irreversible, and injection-aware handling of page content. For developers automating real interfaces without handing over the keys.
Download .md ↓Drive an open-source, provider-agnostic terminal coding agent well — scoped tasks, plan-before-edit, reviewed diffs, and tests as the acceptance gate — so an autonomous coder speeds you up without quietly rewriting things you didn't ask it to. For developers pairing with a terminal AI agent.
Download .md ↓Build, ship, and grow a Telegram bot that meets Telegram's platform rules, adapts to the user's locale and language, and is architected for stability and future expansion. Covers Bot API design, webhook vs long-polling, state and data modelling, localization, rate-limit and error handling, and getting listed. For developers and founders building on Telegram.
Download .md ↓Take generated images and video into a real product pipeline — provenance, licensing, C2PA, naming and versioning, alt text, and the CDS surfaces (AI-made badges, artifact cards) that present synthetic media honestly.
Download .md ↓Design ComfyUI node graphs like production pipelines — planned stages, pinned seeds and versions, VRAM budgets, reproducible workflow JSON — for image and video generation that a team can rerun, not just a lucky pull.
Download .md ↓Plan an AI application on Dify before touching a single node — product logic first, then graph layout, model routing, RAG design, integrations, and the human gates. For designers and developers shipping LLM apps on dify.ai.
Download .md ↓Direct AI image generation like a photographer with a brief, not a slot-machine player — model-agnostic prompt structure, style systems for brand consistency, an iteration ladder, and the honesty rules that keep synthetic imagery legal in a consumer product.
Download .md ↓Direct AI video generation shot by shot — storyboard first, consistency locks, motion budgets, honest editing — plus the disclosure and platform rules that keep synthetic footage out of trouble in a consumer product.
Download .md ↓Checks colour contrast, screen-reader labelling, and full keyboard operation against named WCAG 2.1 AA success criteria, then reviews right-to-left mirroring and long-word / multi-language overflow so a layout survives localization. A combined a11y and i18n pass that cites the specific criterion for each finding and verifies locale rules against the target locale. For designers, front-end, and localization.
Download .md ↓Decide whether Ant Design fits a project before adopting it, then deploy it coherently — design tokens, ConfigProvider theming, component-vs-custom decisions, and adherence checks — so the whole system is used as intended rather than half-applied and drifting. For designers and React developers building data-dense enterprise and admin products.
Download .md ↓Keep a product on the rails of Apple's Human Interface Guidelines from the moment a project decides to target Apple platforms — enforce the non-negotiable rules, earn a premium feel through restraint and native conventions, and protect performance and App Store review readiness. For designers and developers building for iOS, iPadOS, macOS, watchOS, or visionOS.
Download .md ↓Analyse award-winning web design from Awwwards — Site of the Day, Site of the Month, Developer Award — rating design difficulty and build difficulty on a 1–5 star scale, inferring the tech stack, reading the jury criteria, and pulling 3–5 competitor sites relevant to the project you're building. For designers and front-end developers benchmarking ambition against what actually ships.
Download .md ↓Diagnose how mature a product organization's design practice actually is — on a five-level scale from design-as-decoration ("make it look nice") up to design-as-business-strategy — using the published design-maturity model InVision is known for. Answers the real question: are your designers pixel-pushers, or in the room where product decisions get made? For design and product leaders benchmarking design's role.
Download .md ↓An objective read of a design system — no hype, no hate, only what's true. Ed Chen's analysis method: separate fact from judgment, reconstruct the build logic and the context it was made in, then anchor every observation to what the agent's user is actually trying to achieve, and help them get there. For anyone assessing or improving a design system.
Download .md ↓Decomposes a design into named design tokens and fills in every component state a real system needs — default, hover, focus, active, disabled, loading, error, empty — with reviewed Tailwind code. A starting point for a design system that must still pass accessibility and match the project's real config, never colour alone for state. For designers, design-system engineers, and front-end.
Download .md ↓Audits a UI/UX flow for misleading visuals, hidden terms, coerced consent, and other brand-trust-damaging dark patterns, then prescribes high-transparency trust-design fixes measured against a private-banking service standard and international rules (FTC dark-pattern guidance, GDPR consent). A design-level review that cites the relevant rule for each finding and routes legal determinations to counsel. For designers, product, and trust & safety.
Download .md ↓Analyse award-winning work from The FWA (thefwa.com) — FWA of the Day and its higher tiers — rating design and build difficulty on a 1–5 star scale, inferring the tech behind the effects, reading why it was recognised, and pulling 3–5 competitor sites relevant to the project you're building. For designers and developers benchmarking cutting-edge interactive work. Companion to the Awwwards analysis.
Download .md ↓Find and analyse award-winning graphic design from the last five years — The One Show, D&AD, Type Directors Club and peers — then explain each piece's design philosophy and dissect the design logic behind why it won. For designers building visual references and understanding what actually earns recognition.
Download .md ↓Decide whether Material Design (Material 3 / Material You) fits a project before adopting it, then deploy it coherently so the whole system is used as intended rather than half-applied and drifting. Covers fit analysis, token and theme setup, component-vs-custom decisions, and adherence checks that keep an app on-system. For designers and developers building Android, Flutter, or web products on Material.
Download .md ↓For a given industry or domain, quickly assemble the sources worth grounding on — the highest-authority URLs, the peer-reviewed papers, and the official first-party datasets — so a NotebookLM (or any RAG assistant) reasons from primary, current, citable material instead of the open web's noise. For anyone learning a domain fast and honestly.
Download .md ↓Read the user's actual NVIDIA hardware and configure GPU rendering and VRAM budgets that fit it — batch sizes, precision, memory headroom, and the right CUDA/driver assumptions — so a workload runs fully on the GPU instead of OOM-ing or silently spilling to the CPU. For developers running CUDA rendering or ML on NVIDIA GPUs.
Download .md ↓Draws an end-to-end user journey map, annotates the emotion curve and the experience friction at each stage, and turns the worst drop-offs into tested, conversion-lifting recommendations grounded in real research and analytics. A structured read of where users struggle, why, and what to try next — with every inference labelled as such. For UX, product, and CX teams.
Download .md ↓Audit how mature a team's user-experience practice really is — the health of its research process, its design-system maintenance, and its quality control — on a staged UX-maturity scale, in the tradition of the UX consultancy Nielsen Norman Group is known for. A rigorous read of whether UX is absent, siloed, structured, integrated, or user-driven. For product and design leaders raising their UX game.
Download .md ↓Diagnose the operational leverage of a small-to-mid creative, design, or product studio — the ratios that decide whether it makes money — starting with billable utilization (are designers hitting the healthy 60–70% range?) and the non-billable overhead ratio (is admin and sales headcount too heavy?), in the agency-management tradition Agency Management Institute is known for. For studio owners fixing the machine, not the work.
Download .md ↓Assess how exposed a role, a team, a product, or a business model is to being displaced or commoditised by AI — task-level exposure, the defensible core, time horizon, and the reposition-or-reinvest moves — clearly, without hype in either direction. For founders, leaders, and individuals deciding where to invest ahead of AI's advance.
Download .md ↓Spot competition-law risk before it becomes an enforcement action — merger-review exposure, agreements that restrain trade, monopolization and abuse-of-dominance concerns, and information-sharing hazards — so business moves are pressure-tested early, with antitrust counsel making the legal determination.
Download .md ↓Decode how consumers actually decide — using behavioral economics and psychology (heuristics, framing, defaults, social proof, loss aversion) — then validate the effect with real data, so product and marketing choices are grounded in evidence and applied ethically rather than as manipulation.
Download .md ↓Build a defensible market position and the brand equity that commands a premium — a sharp, true positioning statement, differentiation that competitors can't easily copy, and an architecture that holds as the company grows — grounded in real customer and competitor evidence, not adjectives.
Download .md ↓Diagnose whether a company's real resource allocation matches the capabilities its strategy actually needs — then cut the "good enough" and non-differentiating spend to fund the few capabilities that win, in the Capabilities-Driven Strategy and Fit-for-Growth tradition associated with PwC's Strategy&. A structured "prune to the core" body-fitness diagnosis. For leaders whose spend has drifted from their strategy.
Download .md ↓Draft, review, and negotiate commercial and technology agreements — MSAs, SaaS terms, licensing, DPAs, SLAs, NDAs — spotting the clauses that shift risk (indemnity, liability caps, IP ownership, auto-renewal, data rights) and structuring defensible positions, with a licensed attorney making the call.
Download .md ↓A multi-dimensional competitor teardown that compares UI/UX detail, engagement estimates (DAU/MAU), download estimates, business model, and objective public data in one comparison matrix, posts a clear-eyed gap analysis, and reads market trend and latent user need to plan a 6–18 month product strategy. Every traffic and revenue figure is a labelled estimate from public or inferred sources, never a competitor's real internal number. For product leaders, founders, and strategy teams.
Download .md ↓Uses MECE structuring and the 7S framework to produce the written enterprise strategy artifact — a decision blueprint and org-change plan at the standard a Fortune-500 board expects, in the tradition McKinsey is known for. Focuses on the document itself: the argument, the choices, the change plan, and the decision it asks a leader to make. For founders, executives, and strategy leads.
Download .md ↓Stand up a corporate compliance and risk program that actually runs — code of conduct, policies, controls, training, monitoring, whistleblower channels — plus ESG governance and disclosure discipline, so a company manages regulatory and reputational risk by design, with counsel owning the legal determinations.
Download .md ↓Navigate advanced corporate governance — board decision-making and fiduciary duties, shareholder activism and hostile-takeover dynamics, and control transactions — so a board or founder understands the rules of the game and the defensive playbook, with corporate counsel executing the legal steps.
Download .md ↓For a sensitive redesign, security incident, or pricing change, pre-rehearses the negative community sentiment and potential PR crises before launch, then drafts a crisis-response statement and support FAQ that balance legal compliance, genuine sincerity, and a concrete remediation offer. Every statement is a draft for legal and PR review before publishing, never a cleared public communication. For comms leads, founders, and support leads.
Download .md ↓Cost a product team precisely across two scales (5–10 vs 11–50) and across geographies — direct salary, benefits, remote-hardware stipend, SaaS licensing, recruiter fees, and ramp-up friction — then model regional and remote pay premiums and the ROI of a hire before you make it, so scaling grows output, not just headcount. For founders and HR/planning leads sizing a global team.
Download .md ↓Read a product org's health across the dimensions employees actually rate — work-life balance, trust in management, growth, pay satisfaction — surface the cross-team friction (Dev↔PM, PM↔Design) that process and KPI conflicts create, and turn early, consented signals into humane retention action. Aggregate and consented, never covert surveillance of individuals. For HR and founders protecting culture.
Download .md ↓Diagnose a company's health from two directions at once — external customer loyalty (Net Promoter Score) and internal front-line vitality (the founder's-mentality traits of insurgency, front-line obsession, and owner mindset) — to catch "big-company disease" before it shows up in the numbers, in the tradition Bain & Company is known for. For founders and leaders watching for the drift from scrappy to bureaucratic.
Download .md ↓Diagnose the operational and financial health of a digital agency, web-design shop, or app studio — read the time-tracking data to catch the industry's deadliest disease, scope creep (endless unpaid revisions), and fix the pricing and profit structure underneath it, in the agency-operations research tradition Promethean Research is known for. For agency owners whose team is busy but the margins aren't there.
Download .md ↓Benchmark a digital product studio or design agency against its peers on the numbers that matter — compensation levels, project pricing and rates, and profit margins — to see honestly whether the studio is healthy or quietly falling behind the market, in the peer-benchmarking tradition The Bureau of Digital is known for. For studio leaders who need an outside yardstick, not an inside opinion.
Download .md ↓Track competitor product-team talent flow from public signals, then engineer a recruiting funnel that actually screens for skill — attractive, honest job descriptions and real System-Design / scenario interviews that can't be gamed by AI question-banks. For founders and hiring managers scaling a product team.
Download .md ↓Build one comprehensive enterprise health score that synchronises every dimension — financial, spend (software, people, office, legal, admin, product, marketing), people (attrition, satisfaction), market (acceptance, share), and product (lifecycle, iteration, repeat-customer rate) — into a single weighted, transparent diagnostic, in the integrated cross-functional advisory tradition Deloitte is known for. For founders and leadership who want one honest read of company health.
Download .md ↓Diagnose where enterprise value is actually created and lost — a hypothesis-driven, MECE breakdown of the operating model into value drivers, cost-to-serve, and the two or three moves that matter, in the top-down strategy-consulting tradition McKinsey is known for. For founders and operators who need a structured operations diagnosis, not a dashboard.
Download .md ↓Get full visibility into where the money goes and find the savings — categorise total spend across software, people, office, legal, admin, product, and marketing, surface waste, duplication, and maverick spend, and turn it into a prioritised savings plan, in the business-spend-management tradition Coupa is known for. For founders and operators controlling burn.
Download .md ↓Diagnose where an enterprise's operations and technology actually lose efficiency and leak cash — map the real process flows (as executed, not as drawn), find the bottlenecks, rework loops, and waste hiding in ERP, supply-chain, and cross-border operations, in the enterprise-operations and process-mining tradition Deloitte is known for. For leaders whose operation has grown too big to see clearly.
Download .md ↓Assess a company's financial health and ability to meet its obligations with the structured, through-the-cycle credit lens Moody's is known for — leverage, coverage, liquidity, and business risk combined into a transparent internal health grade. An internal analytical rating, never an actual credit rating. For founders, lenders, and operators judging financial resilience.
Download .md ↓Simulates how major competitors might retaliate within one to three months of a new feature or pricing change — price war, fast-follow copying, PR war — and pre-embeds a competitive moat and defensive strategy at the design stage so the move survives the response. Every simulated reaction is a labelled scenario with stated assumptions, not a prediction of what a rival will actually do. For product-strategy leaders, founders, and pricing teams.
Download .md ↓A multi-dimensional market-entry method covering target-market fit, regulatory limits, scaling potential, SWOT, TAM/SAM/SOM sizing, and ARPU / average spend / spend-share economics, paired with a compliance and data-security check (GDPR, CCPA, cross-border transfer limits), an adaptive learning loop that iterates as the business grows, and a concrete landing playbook. All sizing and ARPU figures are estimates with stated assumptions. For founders, growth leads, and international-expansion teams.
Download .md ↓Applies the client-tiering and fiduciary lens publicly associated with private banking to analyse a user's usage data, attitude, engagement, estimated value, and psychology, then builds a high-stickiness bespoke product experience — including a consented Data Gap Strategy that progressively completes an incomplete client profile through painless, privacy-respecting behavioural tagging rather than covert inference. For product, CRM, and wealth/fintech teams.
Download .md ↓Build and defend an intellectual-property position — patents, trade secrets, trademarks, and copyright — and structure technology licensing and cross-border deals, so a company protects what it builds and monetizes it, with patent and IP counsel making the filings and the legal calls.
Download .md ↓Assess where a product or company really stands in its market — addressable market, share and share trajectory, competitive position, and market acceptance — in the market-intelligence tradition S&P Global is known for. Grounded in defined markets and cited data, honest about what's measured versus estimated. For founders and product leaders sizing their standing.
Download .md ↓Design research that yields decisions, not decks — sound experiments and unbiased surveys, plus qualitative depth (interviews, observation) fused with quantitative scale — so you understand real customers from evidence instead of the loudest opinion in the room.
Download .md ↓The quantitative core of modern marketing — customer lifetime value, multi-touch attribution, churn prediction, and dynamic pricing — modelled honestly so budget, retention, and price decisions rest on defensible numbers and disclosed assumptions, not vendor-dashboard vanity.
Download .md ↓Analyse risk across every angle — strategic, financial, operational, market, regulatory, technology, people, and product — in the enterprise-risk-management tradition KPMG is known for: a structured risk register scored by likelihood and impact, mapped to owners and mitigations, with the interdependencies named. For founders and leaders who need risk seen from more than one direction.
Download .md ↓Benchmark your product team's compensation against the market, and watch team-level health signals — consented pulse surveys, aggregate delivery trends, a burnout index — to catch attrition risk early and act on it with concrete, humane action items. Aggregate and consented, never covert individual surveillance. For engineering managers and founders protecting a team.
Download .md ↓Diagnose organizational health — not the financials, but the alignment, execution, and renewal capacity that predict whether performance lasts — using the lens McKinsey is known for through its Organizational Health Index and 7S model. A structured read of how coordinated and capable an organization actually is. For founders and leaders who suspect the problem is the org, not the plan.
Download .md ↓Review a profit-and-loss statement and build a defensible EBITDA bridge with the assurance-first rigor PwC is known for — reconcile the numbers, normalise one-offs honestly, separate GAAP from adjusted, and flag what a real audit would question. For founders and operators preparing financials for a board, a raise, or their own clarity.
Download .md ↓Stress-test a product decision from the other side before you commit — steelman the opposing case, red-team the plan, audit the data the claim rests on, and run a pre-mortem, so weak logic and shaky numbers surface in a review room instead of in the market. Hedged, multi-angle product thinking. For founders and product leaders about to make a bet.
Download .md ↓Precise cost and resource scheduling for product R&D in two modes — General (5–10 person team) and Advanced (11–50 person scale-up) — covering human-resource cost and ramp-up time, engineering-hour scheduling and tech-debt maintenance ratio, training and transformation cost, and Cloud/DevOps FinOps (API fees, GPU and server rental, third-party SaaS licensing, and build-vs-buy outsourcing). Every figure is a model with stated assumptions, verified against current vendor pricing and local rates. For engineering managers, founders, and finance/ops.
Download .md ↓Build a driver-based revenue forecast and FP&A model that a board will trust — grounded in unit economics, tested with scenarios and sensitivities, and honest about its assumptions, in the forward-looking financial-planning tradition EY is known for. For founders and finance leads forecasting revenue for a plan, a raise, or a runway decision.
Download .md ↓Diagnose whether an organization's defenses actually hold — the maturity of its risk controls and compliance posture against cyber threats, regulatory change, and operational failure, structured through the three-lines-of-defense governance model in the risk-and-compliance advisory tradition KPMG is known for. A read of control maturity and defensive mechanisms, not a legal opinion. For leaders who need to know if the guardrails are real.
Download .md ↓Structure and perfect security interests under UCC Article 9, and navigate the mechanics of insolvency — liquidation and reorganization, priority, and creditors' rights — so a lender's or creditor's position is understood and protected, with insolvency counsel making the legal calls.
Download .md ↓Diagnose how ready an organization is to absorb change and shock — geopolitical shifts, technology disruption, regulatory change, and market volatility — and how capable it is of actually executing transformation, in the change-and-resilience advisory tradition EY is known for. A read of organizational antifragility and change-execution capacity, not just a risk list. For leaders facing a change they must survive and lead.
Download .md ↓An adversarial six-persona operating mode for product strategy work — gap diagnosis, stress-tested architecture, empathy refinement, and a quantified delivery lock. Load it into any capable agent to turn a raw idea, brief, or competitor teardown into a defensible product plan.
Download .md ↓Use this skill when drafting or structuring an academic or research paper — literature review, argument scaffolding, or journal formatting and submission. It produces a rigorous, well-cited first draft for the author's own review and verification.
Download .md ↓Use this skill when a request involves putting AI models into production — deploying and serving open-weight or hosted models, cutting inference cost, stabilising data pipelines, or adding reliability and safety backstops. It produces an architecture-and-operations first pass grounded in real builds.
Download .md ↓Use this skill when a request calls for high-craft web design — strong visual direction, scroll-driven motion, and interaction polish — for a marketing site, landing page, portfolio, or product surface. It produces an art-directed first pass that stays accessible and within a performance budget.
Download .md ↓Use this skill when a brand needs a defined voice, a product page needs persuasive copy, or a campaign needs tagline options. It produces a voice system, message-hierarchy copy, and a rationale-backed shortlist ready for review.
Download .md ↓Use this skill when a project needs performant, meaningful web animation on the HTML5 Canvas 2D API — motion design that serves the content, an efficient rendering loop and particle work, and a strict performance and reduced-motion contract. It produces a motion plan and canvas patterns you can build against.
Download .md ↓Use this skill when a request calls for structured career strategy — planning a path to a target role, preparing for a compensation negotiation, or deciding whether to make a career change. Use it to produce a rigorous first-pass analysis quickly while verifying any legal, financial, or contractual specifics with a qualified professional.
Download .md ↓Use this skill to reason through portfolio management and private-wealth planning at CFA Level III depth — capital market expectations and strategic/tactical asset allocation across equity, fixed income, alternatives, and currency; investment policy statements with the full objectives-and-constraints frame for private-wealth and institutional clients (pension, endowment, insurer, sovereign wealth fund, bank); and risk budgeting, behavioral finance, trading/execution, manager due diligence, and GIPS-aware performance evaluation. It structures the analysis the way the curriculum does; it never gives investment advice, makes no market forecast, and routes every actual allocation decision to a licensed adviser who owns suitability.
Download .md ↓Use this skill when a request calls for applied cognitive-behavioral techniques to manage stress, decision pressure, or unhelpful thinking patterns in a high-pressure work setting. Use it to structure a first-pass self-management plan quickly while keeping a qualified professional in the loop for anything clinical.
Download .md ↓Use this skill when a product may collect personal information from children under 13 and you need a fast, structured first pass on COPPA — reviewing a consent flow, mapping data practices, or tracking enforcement. It organises the issues so qualified privacy counsel can act on them.
Download .md ↓Use this skill for the general-management work of running a company or a function — turning strategy into objectives, designing the operating cadence, structuring decisions, and reading organisational health. It produces management artifacts (strategy maps, OKR sets, operating reviews, decision memos) for the leadership team to own — never a substitute for the accountable executives, and not legal, financial, or HR advice.
Download .md ↓Use this skill to work financial accounting and reporting the way a CPA works — analyzing transactions into correct journal entries, preparing financial statements and running the close, and selecting the right US GAAP or IFRS treatment with proper disclosure and internal-control awareness. It structures the accounting; it is never an audit, a tax filing, or a CPA sign-off, and every material or complex treatment routes to a licensed CPA.
Download .md ↓Use this skill to build and interpret EBITDA and the analysis around it — computing EBITDA from the financials, normalising it with defensible add-backs, bridging period-over-period changes, and framing valuation multiples. It produces a sourced, assumption-explicit financial analysis for the CFO/controller and the deal team to review — never audited figures, a valuation opinion, or investment advice.
Download .md ↓Use this skill when auditing the accessibility of a financial-grade frontend (trading, payments, KYC, dashboards) against WCAG and preparing a structured findings record. It produces a rigorous first-pass audit and remediation draft for review by a qualified accessibility auditor.
Download .md ↓Use this skill for people work specific to a regulated fintech — writing role scorecards for compliance-sensitive positions, structuring interviews that test regulatory judgment, and mapping the fit-and-proper / background-screening and accountability considerations a financial firm's hires carry. It produces hiring and people artifacts for HR and compliance to review — never employment-law advice or a regulated fit-and-proper determination.
Download .md ↓Use this skill when a request involves EU/UK GDPR data-protection operations — mapping data flows, reviewing a data-processing agreement, or assessing a cross-border transfer. It produces a rigorous first-pass draft and issue map for a qualified data-protection lawyer or DPO to review.
Download .md ↓Use this skill when a product's funnel needs diagnosing, an A/B test needs designing honestly, or acquisition channels need an ROI comparison. It produces a structured analysis with clear metrics, guardrails, and an honest read-out.
Download .md ↓Use this skill when you need a fast, rigorous first pass on a patent, trademark, or copyright question — capturing an invention, scanning trademark risk, or reviewing a licensing clause. It organises the issues so a qualified IP attorney can act on them.
Download .md ↓Use this skill to prepare a company for a public listing via IPO, direct listing, or SPAC/de-SPAC — configured to the target jurisdiction upfront. It closes and estimates the financials, assembles and matches the required materials and data (financial statements, disclosure, data room), and builds the jurisdiction-specific readiness plan for counsel, auditors, bankers, and the CFO to own — never legal, audit, underwriting, or investment advice, and never a listing or eligibility determination.
Download .md ↓Use this skill when a request involves ISDA derivatives documentation — the Master Agreement, Schedule, Credit Support Annex, or Confirmations — for issue-spotting, term comparison, or negotiation preparation. It produces an organised, well-cited term map and redline-preparation draft for qualified derivatives counsel to review — never a legal conclusion, an executed term, or a netting/enforceability opinion.
Download .md ↓Use this skill when a project needs modern vanilla JavaScript engineered correctly — the language and runtime fundamentals people get subtly wrong, DOM and browser work without a framework, and the testing, performance, and security discipline that keeps it shippable. It produces concrete patterns and a review you can act on.
Download .md ↓Use this skill to help someone learn a subject or read difficult material more effectively — diagnosing where they're stuck, building an evidence-based study system (spaced repetition, retrieval practice, interleaving, the Feynman technique), and improving reading comprehension and retention. For designing a study plan for an exam or new field, fixing a "I read it but nothing sticks" problem, or coaching a durable reading habit. Not clinical assessment — real learning disabilities or attention conditions route to a qualified specialist.
Download .md ↓Use this skill when a request involves preparing a broker, exchange, or company for a listing or regulatory admission review — generating the review checklist, tracking gaps with reminders, and cross-referencing prior cases. It produces an organised, well-cited readiness checklist and gap tracker for counsel, sponsors, and the deal team to review — never a listing opinion or an eligibility determination.
Download .md ↓Use this skill to turn a tangled problem, question, or argument into a clear, navigable structure — decomposing it into a MECE issue tree or first-principles components, mapping what depends on what and what is assumed, and building the synthesis top-down (Minto pyramid style) so the conclusion and its support are legible. It organises reasoning; it does not audit its truth.
Download .md ↓Use this skill when a real problem needs to be turned into a quantitative model and analysed rigorously — stating assumptions and choosing a deterministic or stochastic form, solving and calibrating it honestly to data, and then validating it with sensitivity analysis, uncertainty quantification, and out-of-sample checks so a model's outputs are read as a simplification, not as truth.
Download .md ↓Use this skill when a request involves KYC/AML compliance across more than one jurisdiction — mapping the requirements per country, generating a gap/comparison report against a program, or shaping how findings feed an existing compliance system. It produces an organised, well-cited requirements matrix and gap report for a qualified MLRO / compliance officer to review — never a compliance determination or a filing.
Download .md ↓Use this skill to design, build, and harden workflow automations in n8n — mapping a manual process to triggers and nodes, wiring APIs and data transforms, and making the result reliable, secure, and affordable to run. For automating a repetitive multi-tool process, debugging a flaky workflow, or reviewing an automation before it touches production data. Not official n8n support, and never a licence to move money, delete data, or act irreversibly without a human in the loop.
Download .md ↓Use this skill when a claim, argument, analysis, or piece of copy needs to be checked for objectivity and logical soundness before it ships — separating claim from evidence, naming the inference gaps and fallacies, stress-testing with the strongest counter-argument, and calibrating confidence to what the evidence actually supports.
Download .md ↓Use this skill when a request involves cutting the output-token cost of an AI workflow without losing quality — tightening prompts and system messages, restructuring what the model is asked to emit, and putting cost governance around an agent or pipeline. It produces a measured before/after with the levers pulled and a guardrail so quality is verified, not assumed.
Download .md ↓Use this skill when you need a fast, structured prior-art search and patent-landscape scan before deciding to file, invest, or engage a patent attorney. It produces a documented search log and an issue map organised against the statutory elements — never an opinion.
Download .md ↓Use this skill to turn real research into decision-useful personas — evidence-grounded archetypes with goals, contexts, and jobs-to-be-done that a product or design team can actually build against. For synthesising interviews, surveys, support tickets, or analytics into a small set of personas, pressure-testing existing personas for bias, or turning a persona into journeys and scenarios. Not a substitute for talking to real users, and never a licence to invent a "typical user" from assumptions.
Download .md ↓Use this skill when sizing demand, comparing competitor pricing, or scoring product-selection opportunities for a merchandising or product decision. It produces a structured, directional first pass to be validated with primary data before any commitment.
Download .md ↓Use this skill when a request calls for search-engine strategy — keyword opportunity analysis, a technical SEO audit, or a content blueprint. It turns a vague "we need to rank / grow organic" ask into a structured, prioritised first pass.
Download .md ↓Use this skill when a Solidity/EVM smart contract needs a fast, structured first-pass security review before a professional audit. It maps likely vulnerability classes, drafts gas notes, and produces an audit-report skeleton for a qualified auditor to verify.
Download .md ↓Use this skill to help someone think their way to their own answer rather than be handed one — through Socratic questioning, deliberate perspective-shifts, and a concrete action trigger at the end. For coaching a decision, unsticking a stuck team, running a reflective 1:1, or pressure-testing an idea by drawing out the thinker's own reasoning.
Download .md ↓Use this skill when a TypeScript codebase needs stronger type modelling, a safe refactor of loose or JavaScript code, or an architecture review of a frontend app. It produces a concrete, staged plan and typed patterns you can adopt incrementally.
Download .md ↓Use this skill when a request involves US federal securities-law issue-spotting — registration vs exemption, disclosure and periodic-reporting obligations, or broker-dealer / investment-adviser conduct rules. It produces an organised, well-cited issue map for licensed securities counsel to review — never a legal conclusion, a filing position, or an offering decision.
Download .md ↓Use this skill when a request involves US federal or state tax issue-spotting, entity-treatment comparison, or cross-border reporting exposure. It produces an organised, well-cited issue map for a licensed CPA, tax attorney, or Enrolled Agent to review — never a filing position.
Download .md ↓Use this skill to write director-grade prompts for AI video generation, scaled to the length you actually need — a 10-second hero clip, a 1-minute sequence, or a 30-minute program. It brings shot language (framing, lens, motion, light, pacing) and, for longer pieces, decomposition into shots and scenes with continuity control. For turning an idea into a shot-specification a model can render, or planning a longer piece as a shot list. Not a guarantee of output, and never for cloning real people or copyrighted characters.
Download .md ↓Use this skill when a token economy, launch roadmap, or DAO governance model needs a structured first-pass design and risk map before legal, financial, and tax counsel review. It drafts the model and flags where a design may trigger securities or other regulation — never a conclusion.
Download .md ↓Use this skill when you need a structured, self-management first pass on a work-stress, conflict, or boundary situation — recognising burnout signals, preparing a difficult conversation, or setting a boundary. It organises the issues and routes clinical, safety, or HR/legal matters to qualified professionals.
Download .md ↓No skill matches — clear the search or switch kits.
Every card downloads a single plain-Markdown file. No account, no build step, no tracking.
Add it as a Claude Skill, paste it into a chat, or drop it into a local model's context. The consumer kit also binds to live machine contracts — see the MCP layer.
Each skill states what it is not and when to escalate to a human or a licensed professional. If a skill's guardrail conflicts with what you want, the guardrail wins — that is the design.
The skills above are methods — how to do a piece of work. A separate set of files does the opposite: it describes a mind completely enough that a model can argue from inside it. Same artifact class as everything on this page — one self-contained Markdown file, no runtime, no account — built to one repeatable four-part specification across three very different traditions.
Fifteen statesmen, poets and philosophers rebuilt from their primary writings as decision protocols, seated with their historical opponents.
20 Athenaeum Western philosophy & artTen philosophers and ten artists, Socrates to Dalí, each a way of seeing with five fixed evaluation dimensions and a critique protocol.
27 Juris-Counsel Adversarial law · fictional compositesAdvocates, the bench, and specialist practice counsel — anchored to real statute and precedent, and explicit that none of it is legal advice.