Progressive-disclosure instructions an agent loads on demand. Cheap until invoked, which is why the per-session figure on each card is usually two digits.
Create, review, audit, migrate, or update scoped AGENTS.md files with clear rules, verified commands, action limits, instruction order, and checks that prove the work is done. Use also when verified, durable repository guidance emerges during authorized work and should be captured at the narrowest applicable scope.
Fetch daily arXiv announcements (cs.RO), filter by research relevance, generate a structured markdown digest, and sync matched papers to a Zotero collection with arXiv PDFs attached.
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
Use TreeWork Manual when the user asks for it, or when the Agent judges that organizing ongoing, multi-part work as a persistent Markdown Tree would improve direction, continuity, recovery, or handoff. Adapt the Tree's depth and document detail to the user's needs, domain, and situation rather than a fixed threshold.
Deep multi-model council with research, adversarial debate, sequential reasoning, and final synthesis. For architecture decisions and complex tradeoffs.
Reviews an Agent Skills skill against a strict, weighted quality rubric and produces a score (0-100%) with actionable issues. Use when asked to review, score, audit, or evaluate a skill or to check quality. Triggers on: "review this skill", "score the skill", "audit skill quality", "is this skill ready to merge". Do…
A skill for creating Excalidraw presentations and diagrams. Excalidraw is a drawing tool for making visual explanations, such as process flows, architecture diagrams, and teaching materials.
A project-initialisation skill that creates a lightweight structure of guides for AI coding agents. It maps project areas and adds sections such as contents, locations, commands, working methods, and cautions.
Build image2-first classroom PPT page images from a topic, source files, reference PPT style, and iterative user feedback. Use when Codex needs to create image2-ready per-page prompts, generate polished PPT page images, QA a contact sheet, insert the final images into a PPTX wrapper, and write a timed presentation…
Search for music tracks using natural language prompts or video content via the Harmix AI API. Use when the user asks to find music, search for songs, discover tracks, needs music recommendations based on mood/genre/tempo/scene, or wants to find music for a video, video soundtrack, or music that matches video content.
Use when writing C# / VB.NET / F# code that reads, writes, manipulates, or exports Excel (.xlsx / .xls) files, generates PDF or HTML from Excel, or produces data-driven reports with FlexCel Studio for .NET (TMS Software). Triggers include Excel/xlsx from C#, XlsFile, FlexCelReport, FlexCelPdfExport, FlexCelHtmlExport…
Initiate the orchestrator-worker cost model for this session. The main-loop model (expensive, metered) orchestrates, specs, verifies, and validates only; ALL drafting and mechanical work (code, docs, tests, sweeps, research) runs in Opus 4.8 subagents billed to plan allowance. Invoke at session start or whenever token…
Use when a user asks what Atlas Flight Booking can do, wants to authorize Atlas, search or compare flights using exact or flexible dates and time preferences, verify a current fare, choose baggage or seats, create and pay for an order, or check ticketing status.
CEO/founder-mode plan review. Rethink the problem, find the 10-star product, challenge premises, expand scope when it creates a better product. Three modes: SCOPE EXPANSION (dream big), HOLD SCOPE (maximum rigor), SCOPE REDUCTION (strip to essentials). Originally from Garry Tan's gstack…
Use when a user wants to plan and implement a new feature or bug fix from scratch. Triggers on requests like "I want to build X", "add a feature for Y", "fix Z", or when the user invokes /plan-and-build. Guides through clarifying questions, breaks work into small features, sets up a tasks folder, implements each…
Use when designing, reviewing, refactoring, or enforcing Angular application architecture with isolated lazy features, eager core/layout, reusable ui/pattern architecture building blocks, one-way dependency graphs, eslint-plugin-boundaries rules, or folder placement decisions. This skill takes precedence over…
Audit an app, website, metadata, SDKs, and marketing for privacy, security, AI transparency, consumer claims, subscriptions, content rights, user-generated content, children, health, financial, export, and platform-policy risks. Use before TestFlight, submission, release, paid marketing, or after adding tracking…
Cryptographic compliance verification. Sign security tool output into verifiable CPOEs (JWT-VC), verify vendor proofs via trust.txt, detect drift with diff, and assess third-party risk. Use when the user mentions compliance proofs, CPOE, trust.txt, SCITT, vendor assessment, GRC evidence, or compliance drift.
A coding-agent skill for generating PowerPoint presentations, with instructions and documentation for the NanoBanana PPT generator.
★not rated 14 todayB0 tokens
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: