AI & ML

fradser Bundle

Creates clean, conventional git commits using standard git. Use when the user asks to "commit", "git commit", "create commit", or wants to commit staged or unstaged changes.

fradser Bundle

Creates clean conventional git commits using standard git and pushes changes to the remote repository. Use when the user asks to "commit and push", "push changes", or complete a commit followed by git push.

mmkkkkzz Bundle

Log in and verify Codex MCP servers on this Mac. Use when the user asks to authenticate, reauthenticate, reconnect, or make all configured Codex MCP servers usable, including OAuth MCPs such as Notion, Sentry, Supabase, Travel Planner, and Vercel, plus local HTTP MCP health checks.

mmkkkkzz Bundle

日本語の技術文書・書籍原稿の文章規範。整形（一文一行、引用ブロック、脚注、コラム記法）、段落と論証の構成（パラグラフライティング）、論証の厳密さ（ツッコミどころの除去）、読み手の負荷の管理、視点と語り、演出の抑制、LLM っぽい空句の禁止、冗長の排除を定める。日本語で技術書の章、草稿、記事、解説文を書くとき、または推敲・リライトするときに使用する。

mmkkkkzz Bundle

Resolve actionable GitHub PR review findings end to end. Use when the user asks Codex to address PR review feedback, review findings, session/prompt-supplied findings, unresolved GitHub PR review comments, or "対応すべきものは根本対応してpush、不要なものは理由付きで報告、対応したPRレビューコメントはresolve" style requests.

pramoddutta Skill

Builds a Playwright Page Object Model class from a page or URL. Use when an SDET says "make a page object for the login page", "build a POM for the dashboard", "extract locators into a page class", or wants to refactor inline selectors into a reusable class. Produces locators-as-methods (getByTestId/ getByRole), action methods, and a static PATH — a draft to review.

intel Bundle

Recommend a vLLM-XPU deployment config (quant, KV dtype, DP/TP, max concurrency, max context) for a Hugging Face decoder-only LLM on Intel Arc B-series GPUs using roofline math against published hardware specs. Experimental; predictions are physics-bounded ranges, not measured throughput. Use after xpu-discover and before vllm-xpu-run.

ruvnet Skill

Quantum-resistant, self-learning version control for AI agents with ReasoningBank intelligence and multi-agent coordination

ruvnet Skill

MCP server optimization and transport layer enhancement for claude-flow v3. Implements connection pooling, load balancing, tool registry optimization, and performance monitoring for sub-100ms response times.

ruvnet Skill

15-agent hierarchical mesh coordination for v3 implementation. Orchestrates parallel execution across security, core, and integration domains following 10 ADRs with 14-week timeline.

modem-dev Skill

Core agent-browser usage guide. Read this before running any agent-browser commands. Covers the snapshot-and-ref workflow, navigating pages, interacting with elements (click, fill, type, select), extracting text and data, taking screenshots, managing tabs, handling forms and auth, waiting for content, running multiple browser sessions in parallel, and troubleshooting common failures. Use when the user asks to interact with a website, fill a form, click something, extract data, take a screenshot, log into a site, test a web app, or automate any browser task.

kreuzberg-dev Skill

Use tree-sitter-language-pack mcp reference documentation.

kreuzberg-dev Skill

Use when the user wants to split source code into chunks for an LLM context window without breaking syntax mid-construct. Covers `ts-pack process --chunk-size`, why syntax-aware splits beat fixed-byte splits, picking a size, and the chunk JSON shape.

kreuzberg-dev Skill

Use when parsing source, extracting code structure, or detecting a language through the tree-sitter-language-pack MCP server's tools, rather than shelling out to the ts-pack CLI. Covers the tool surface, the auto-installing launcher, and when MCP beats the CLI or SDK.

kreuzberg-dev Skill

Use when parsing source, extracting code structure, or detecting a language through the tree-sitter-language-pack MCP server's tools, rather than shelling out to the ts-pack CLI. Covers the tool surface, the auto-installing launcher, and when MCP beats the CLI or SDK.

yurifrl Skill

Semantic code search using RAG and vector embeddings. Use when you need to find code by concept/functionality rather than exact text match, discover implementation examples, or understand where specific concepts are used across projects.

stacksjs Skill

Use when creating event listeners in app/Listeners/ — the listener file structure, registering listeners in app/Events.ts, the listener-to-action mapping pattern, CLI event listeners in Console.ts, or debugging listener execution. For the event system API (dispatch, listen, emitter, model events), see stacks-events.

carlosferorduna Skill

Testing conventions - Jest framework, unit tests alongside the code they cover, integration tests in a separate test/ directory, and ephemeral agent tests removed when the work is done. Load when writing or modifying tests.

carlosferorduna Skill

Always-on engineering contract for this project - language rules, security, architecture, package manager, timestamps, logging, and agent workflow, plus an index of the stack skills. Load at the start of any coding, review, or commit task; load the relevant stack skill on demand.

kyrie66nb Skill

Resolve references to the currently open or selected Simulink context ("this model", "this subsystem", "this block", "selected blocks") and find blocks in an open model. Use whenever the user refers to "this model", "this block", "this subsystem", "selected blocks", or asks to find specific blocks in an already-open model. After resolving context, hand off to the appropriate Simulink skill for edits, simulation, or testing.

itspecialist111 Bundle

Jurisdiction-aware legal document redlining agent: ingests customer-facing legal documents (NDAs, MSAs, DPAs, T&Cs, SOWs), segments them into clauses, rewrites each clause to align with internal policy positions and regional legal requirements, produces a fully redlined document plus a structured audit pack with change-by-change rationale and citations. Supports UK, EU, US, and custom jurisdictions. Use when user says "redline this contract", "review against our playbook", "localize this agreement", "adapt for EU", "adapt for UK", "jurisdiction review", "policy-aligned redline", "clause review", "contract redline", "playbook review", "mark up this agreement", "regional adaptation", "legal playbook", "NDA review", "MSA review", "DPA review", "T&Cs review", "standard positions review", "contract markup with audit trail", or needs jurisdiction-specific, policy-aligned contract redlining with full auditability.

itspecialist111 Bundle

rfp-gates

itspecialist111 Bundle

Automated ingestion, parsing, and classification of incoming Request-for-Proposal (RFP), RFI, RFQ, tender, bid, and security-questionnaire documents. This skill extracts every question, classifies it by team (Security, Technical, Commercial, Company, General), detects mandatory vs optional items, extracts buyer metadata (deadline, submission format, evaluation criteria, page limits), and produces the structured task list that every downstream RFP skill consumes. Use when the user says any of: - "a new RFP came in, can you parse it" - "parse this RFP" - "intake this proposal" - "classify RFP questions" - "extract RFP questions from this document" - "log this RFP in our system" - "process this tender" - "ingest this bid document" - "break down this RFP by team" - "what's in this RFP, give me the summary" - "triage this security questionnaire" - "set up the task list for this bid" Do NOT use for: making the bid / no-bid decision (route to rfp-fit-assessment), drafting answer content (rfp-respond), running qualit

itspecialist111 Bundle

Step 5 of the RFP agent swarm. Surfaces only the flagged items that need human attention, captures every reviewer correction with a structured reason, and feeds the learning loop that improves the answer bank for future RFPs.

itspecialist111 Bundle

Drafts first-pass answers to every RFP question using a strict retrieval-first pattern, routes questions to Security / Technical / Commercial teams in parallel, and stamps every response with a confidence tier and provenance before handing off to quality gates.

itspecialist111 Bundle

Manage the Contoso RFP answer bank — search, add/update, sync Loopio exports, merge reviewer corrections. This is the retrieval substrate that every other rfp-* skill depends on.

itspecialist111 Bundle

Step 2 of the RFP Agent Swarm workflow. Produces a Go/No-Go recommendation MEMO for a freshly intaked RFP by running a weighted qualification scorecard across seven dimensions (KB match, technical fit, commercial fit, competitive position, strategic alignment, resource availability, deadline feasibility) and surfacing evidence, risks, and kill-criteria. The human ALWAYS makes the final bid/no-bid call — this skill only produces data and analysis. Use when the user says things like: - "should we bid on this" - "go/no-go analysis" - "fit score this RFP" - "qualify this RFP" - "bid decision" - "assess this RFP" - "should we respond" - "run fit assessment" - "score this opportunity" - "is this RFP worth pursuing" - "qualify [buyer name] RFP" - "build the Go/No-Go memo for leadership" Do NOT use for: parsing raw RFP documents (use rfp-intake), drafting the actual answers (use rfp-respond), running quality gates on drafts (use rfp-gates), assembling the final response document (use rfp-assemble), or maintaining the

qovery Bundle

Takes a brand-new user from nothing to a working Qovery account and first organization — checks the Qovery CLI is installed (installs it if missing), signs them in with `qovery auth --headless` (which creates the account on first OAuth login), then creates an organization via the API, enriches it from the company website, records the sign-up for tracking and lead qualification (tagged as a CLI/agent sign-up), and sets the CLI context. Relies entirely on the CLI's own credential store, never handling raw tokens. Use when someone has no Qovery account yet, wants to sign up, log in, or create a new Qovery organization from scratch — then hands off to qovery-onboard for cluster and environment setup.

qovery Bundle

Creates a scoped Qovery API Policy Token from a user's intent — authors a least-privilege Open Policy Agent (Rego) policy, tests it locally with OPA and live against the API, creates the token, and verifies it allows exactly what the user wants and denies everything else. Use when the user wants a restricted, least-privilege, or agent-scoped Qovery API token, an OPA/Rego policy token, or wants to lock a token down to specific environments, services, or actions.

qovery Bundle

Diagnoses and fixes deployment failures, application crashes, build errors, connectivity problems, stuck deployments, and cluster issues on Qovery. Uses a systematic 8-layer diagnosis with MCP Server integration, CLI, and API, and generates runbooks for recurring issues. Use when the user reports a Qovery deployment that is failing, broken, stuck, or crashing. (For slow deployments use qovery-speedup; for cost optimization use qovery-optimize.)

suda777 Bundle

Create or update a concise, verified project AGENTS.md for code, consulting, materials, or mixed projects, then audit the result for truth, scope, duplication, conflicts, privacy, and maintainability. Use when the user asks to create, write, generate, update, refresh, maintain, repair, audit, or reorganize a project AGENTS.md or project agent guide. Do not use for the global ~/.codex/AGENTS.md, ordinary README editing, one-off task plans, handoffs, or generic documentation that is not agent guidance.

typefox Bundle

Run the eval loop that measures whether a skill actually improves an agent's output — design test cases, run them with the skill and against a baseline, grade against assertions, aggregate a benchmark, analyze patterns, and iterate. Use when the user wants to evaluate or test a skill, check whether a skill helps, run skill evals, benchmark a skill against a baseline, grade skill outputs, or iterate on a skill from eval results. Builds on the skill-creator skill for the mechanics; stop if skill-creator is unavailable.

typefox Bundle

Make a code repository agent-ready with state-of-the-art AX (Agent Experience): audit and set up AGENTS.md / CLAUDE.md and agent docs, wire verification sensors with self-correction messages, and build docs-as-memory (architecture docs, ADRs, exec plans). Use when the user wants to make a repo agent-ready or AI-friendly, onboard coding agents, create/review/improve AGENTS.md, CLAUDE.md, or other agent instruction files, set up a project for agent-first development, asks why an agent keeps repeating a mistake, asks any AX question (general or project-specific), or mentions "harness engineering", "agent readiness", "context engineering" (its repo-side slice), or "AX". Not for building agent runtimes or orchestration code, nor for authoring a single agent skill — use skill-creator for that.

typefox Skill

Work with the Domain Model DSL (.dmodel) for describing data models — datatypes, entities, and their features. Use when the user wants to create or edit a domain model in this DSL.

typefox Skill

Read, write, and modify Domain Model DSL files (.dmodel) — datatypes, entities with inheritance and multi-valued features, and nested packages with qualified cross-references. Use when the user wants to create, edit, explain, or debug a domain model in this DSL, including resolving references across packages.

yunchengliu Bundle

Guide modern, reliable PyTorch mechanics for single-machine workloads: modules, datasets, dataloaders, train/eval/inference loops, devices, dtypes, checkpoints, AMP, torch.compile, and basic single-node DDP. Use for PyTorch implementation choices, not general experiment planning or cluster design.