Total 58,118 skills
Showing 12 of 58118 skills
Diagnose a product against Ivan Zamesin's Next Move Theory / Advanced Jobs To Be Done methodology (distinct from generic Christensen JTBD) and route to the next move. A conversational, chat-first skill — through up to ~15 adaptive questions it challenges the goal you walked in with (climbing your business-Job graph to find a higher-leverage growth point), then surfaces EVERYTHING the methodology can find: all the risks (every weak node on the chain to profit, each traced to its upstream cause), all the growth points (kill a Job, Previous/Next Job, climb a level, adjacent-segment Small Jobs, underserved success criteria), and the risky assumptions hiding inside your current initiatives (RAT). It is comprehensive on findings and focused on action — it lists them all, then says which one to tackle first and which existing skill executes it (nmt-market-research, nmt-craft-value- proposition, nmt-product-requirements, nmt-craft-go-to-market, nmt-chat). The front door for LIVE products the way nmt-market-research is the front door for new ideas. Use when the user says "diagnose my product", "what should I do next", "a metric dropped", "where are my risks / growth points", "where is this breaking", or is unsure what to work on. Chat-first: writes nothing unless asked; recommends the next skill but the user launches it. Plain language; defaults to English.
Génère une documentation d'API claire et structurée à partir de code, routes ou descriptions. À utiliser quand l'utilisateur veut documenter une API REST, GraphQL ou des endpoints. Se déclenche aussi avec "documente mon API", "swagger", "endpoint documentation", "API docs", ou quand l'utilisateur montre des routes/controllers. Also triggers on "document my API", "generate API docs", "OpenAPI documentation".
Bounded resolver for confirmed P0/P1 findings — for each: fix, verify the fix resolved it, up to 2 attempts, then escalate. TRIGGER when: a review or fan-out skill needs confirmed findings driven to resolution; user says 'fix these findings' or 'resolve the P0s'. Scoped to per-finding resolution; regressions are the caller's job.
Create distinctive, production-grade frontend interfaces with strong hierarchy, thoughtful systems, and polished implementation that avoid generic AI aesthetics. Use when the user wants to build or redesign web pages, flows, components, or full app surfaces, or when another better-web-ui skill needs shared project design context before other better-web-ui skills.
Coordinate a high-fidelity Figma-to-code workflow for screens, layouts, drawers, dialogs, or component groups. Use when Codex receives one or more Figma URLs or node IDs and must decide how to extract exact design intent, split work into reusable components, route implementation safely, and keep conversion and apply work aligned with the local design system and verification constraints.
Run four parallel read-only subagents that each review the same diff from a different lens — security, performance, correctness, and readability — then merge findings into one report. Use before merging large or risky PRs.
Sizes a Qdrant deployment before it is provisioned. Use when someone asks 'how much RAM do I need', 'how many nodes', 'how big should my cluster be', 'sizing', 'capacity planning', 'will N vectors fit', 'what instance type should I pick', or gives a vector count and dimensions and asks what to provision. Also use when an existing estimate needs checking before hardware or a cluster tier is bought.
Complete creator operations (publish / update / preview / query / troubleshoot) on Bilibili's Toy platform via Toy CLI. Trigger words: Toy Publish, Publish Toy, toy create, Upload Toy, Toy Update, toy update, Change Toy Cover, Change Toy Password, Toy Preview, Toy Submit for Review, My Toy, toy mylist, Toy List, Toy Statistics, toy stats, Toy PV/UV, Toy Login, toy login, Toy Publish History, toy history, toy cli, Toy White Screen, Toy 404, Toy Page Unavailable, Toy Resource Path, Toy slug. Use this skill when users need to publish or update local frontend projects/HTML/zip to the Toy platform, view their own Toys, check access statistics, troubleshoot white screens/resource 404 issues after publishing, or manage local publish records.
Upgrade an existing website or app to premium quality without breaking functionality — audit the current design, identify generic/AI tells, then apply taste and system rules surgically. Use when the user wants to improve, modernize, polish, or "make better" an existing UI/codebase.
Produces human-sounding text that avoids detectable AI writing patterns. Activates on any writing task — tweets, emails, articles, bios, captions, reports, copy, messages, LinkedIn posts, cover letters, README files, or any content where the output must not read as AI-generated. Enforces banned vocabulary, structural variety, punctuation discipline, accuracy rules, and voice calibration. Use when the user says "write," "draft," "rewrite," "make this sound human," "anti-slop," "not AI," or any variation of wanting authentic-sounding output.
Used for authorized reverse engineering of custom binary protocols, Protobuf/gRPC, WebSocket frames, and PCAP-driven protocol recovery.
N-day Patch Diff to Exploit. Reverse-engineer vulnerability points from vendor-released patches, write PoCs, and develop usable attack modules. Applicable scenarios: Known CVE number but only patches available without PoC; SRC/red teams need to target unpatched assets; N-day weaponization; Patch Tuesday follow-up. Core method: Obtain before/after binaries → align symbols → binary diff → reverse-engineer bug class by examining newly added security checks → write PoC to trigger the vulnerability. Trigger keywords: N-day, Nday, patch diff, patch diff, patch tuesday, 1day, binary diff vulnerability, bindiff exploitation, ghidriff, Diaphora, patch analysis, CVE reproduction, vulnerability restoration, patch reverse-engineering, N-day weaponization.