AI & ML

necmttn Skill

Model-routing orchestration for any expensive frontier model (Fable, Opus, GPT-5.x) - the main model keeps judgment and Q&A review, mechanical subagent dispatches carry an explicit cheaper model, and ax measures whether the routing actually worked. Use when orchestrating codebase-heavy or token-heavy work with subagents, when dispatching Agent tasks without a model, when the user says "route to cheaper models", "efficient dispatch", "optimize model spend", or asks where their token spend goes. Pairs with the route-dispatch hook (deterministic backstop) and `ax dispatches` (evidence). Do NOT fire on single-shot questions or tiny tasks with no dispatching.

cartography-cncf Bundle

Define a new node schema under cartography/models/MODULE_NAME/, including required properties, sub-resource relationships, extra labels, conditional labels, scoped cleanup, and one-to-many transforms. Use when the user asks to add a node type, model a new resource, configure extra Neo4j labels (Identity, Asset, UserAccount, Tenant), or wire scoped vs global cleanup.

cartography-cncf Bundle

Author a new Cartography intel module end-to-end (entry point, sync GET/TRANSFORM/LOAD/CLEANUP, declarative data model, integration test, schema docs). Use when the user asks to add a new provider, integration, intel module, or service ingestion to Cartography (e.g. "add a new module for service X", "integrate ServiceY", "create a sync for Z API").

cartography-cncf Skill

Convert a legacy handwritten-Cypher Cartography sync (`load_*` / `cleanup_*` JSON jobs) into the modern declarative data model (`load()`, `GraphJob.from_node_schema()`). Use when the user asks to refactor, modernise, migrate, or "clean up" a legacy intel module, or to remove a `cleanup/*.json` job tied to an old `MERGE` query.

cartography-cncf Bundle

Define a `CartographyRelSchema` (standard relationship), one-to-many edge, or `MatchLink` connecting existing nodes. Use when the user asks to add a relationship, link nodes, set a `RESOURCE` / `MEMBER_OF` / `ASSOCIATED_WITH` edge, share a node across modules, or model a composite node from two intel sources.

kesslerio Skill

Implementation/review workflow for approved plans. Route ACP-first, then CLI fallback. Use after explicit APPROVE.

confluentinc Skill

Use when the user asks about Model Context Protocol details (protocol spec, transports, authorization, tools/resources/prompts, sampling/elicitation/roots, server/client concepts, extensions, or SDK usage). Triggers on phrases like "MCP spec", "MCP transport", "streamable HTTP", "MCP authorization", "list_tools", "initialize handshake", "MCP SDK", "@modelcontextprotocol/sdk", or "how does MCP X".

confluentinc Skill

Reviews pull requests for the Confluent MCP server. Use when reviewing PRs, doing self-review before sharing with the team, or when the user mentions "review PR", "help with PR", "review changes", "self-review", "review local changes", or "check my PR". Focuses on MCP tool wiring, OpenAPI/type coupling, ESM stubbability, transport security, and project-specific patterns.

chubbyguan Bundle

Obsidian 知识库全生命周期管理：三层架构、素材入库(ABC分级)、健康检查、GBrain/GraphRAG/LLM Wiki 三件套集成、目录整理

learning-bayesian-statistics Bundle

Production-grade Bayesian causal inference with PyMC, CausalPy, and DoWhy. Enforces DAG-first thinking, mandatory user checkpoints for assumptions, design-specific refutation, and defensible reporting with causal language guardrails. Trigger on: causal inference, causal effect estimation, treatment effects, counterfactuals, difference-in-differences (DiD), synthetic control, regression discontinuity (RDD), interrupted time series (ITS), instrumental variables (IV), propensity scores, DAGs, causal graphs, confounders, backdoor criterion, do-calculus, interventional distributions, pm.do(), pm.observe(), CausalPy, DoWhy, mediation analysis, refutation, sensitivity analysis, parallel trends, placebo tests, or any question of the form "does X cause Y" or "what is the effect of X on Y."

learning-bayesian-statistics Bundle

Opinionated Bayesian modeling workflow with PyMC and ArviZ. Contains critical guardrails (nutpie sampler, prior/posterior predictive checks, LOO-PIT calibration, prior sensitivity checks, 94% HDI, non-centered parameterizations, reproducible seeds) that agents won't apply unprompted — always consult before writing Bayesian model code. Trigger on: building probabilistic/Bayesian models, prior elicitation, MCMC inference, convergence diagnostics (divergences, R-hat, ESS), model comparison (LOO-CV, ELPD, stacking weights), hierarchical/multilevel models, count regressions, logistic regression with uncertainty, prior sensitivity analysis, reporting Bayesian results, or mentions of PyMC, ArviZ, InferenceData, credible intervals, posterior distributions, shrinkage, uncertainty quantification. Also trigger for model comparison, diagnosing sampling problems, choosing priors, or presenting stats to non-technical audiences.

learning-bayesian-statistics Bundle

Opinionated amortized Bayesian workflow with BayesFlow for simulation-based inference (SBI). Contains critical guardrails that agents will usually not apply unprompted — always consult before writing BayesFlow code. Trigger on: simulation-based inference, amortized inference, approximate Bayesian inference, BayesFlow, neural posterior estimation, posterior amortization, simulator design, prior design for SBI, offline/online simulation pipelines, uncertainty quantification from simulators, structured data encoders (sets, time series, images), or mentions of BasicWorkflow, fit_online, fit_offline, fit_disk, flow matching, diffusion model, consistency model, normalizing flow, summary networks, adapters, or simulation budgets.

tswicegood Bundle

Evaluate prompts, skills, and Claude Code agents for correctness and optimization against the latest Claude best practices. Use when the user asks to "evaluate," "review," "audit," "optimize," "modernize," or "upgrade" a prompt, system prompt, SKILL.md, agent definition, or CLAUDE.md, or says things like "I wrote this for Opus 4, is it still good on 4.7?" Researches current official guidance first, then community reports of changes, then produces a prioritized findings report and (on request) a rewritten artifact.

skyvern-ai Skill

Run smoke tests against a deployed or local app based on your git diff. Each test uses Skyvern browser tools (navigate, act, validate, screenshot) with Chrome DevTools MCP as fallback. Posts screenshot evidence as PR comments.

mistralai Skill

Async and concurrency patterns for Mistral Vibe. Use when working with asyncio, the agent loop, Textual TUI threading, HTTP clients, streaming surfaces, or core-to-TUI communication.

mistralai Skill

Write or refactor Mistral Vibe tests with proper decoupling. Use when adding behavior coverage, testing ports/adapters, replacing brittle mocks, creating fakes, adding characterization tests before refactors, or changing tests under tests/ for vibe/core, vibe/cli, vibe/acp, tools, config, sessions, skills, hooks, MCP, or setup.

mistralai Skill

Add or modify a feature in the Mistral Vibe Python CLI while following the app architecture ADRs. Use when working in vibe/core, vibe/cli, vibe/acp, vibe/setup, built-in tools, config, sessions, skills, hooks, MCP, connectors, or other Vibe feature code.

d-kimuson Skill

Run the claude-code-viewer release flow end-to-end. Use when the user asks to release patch, minor, major, beta, or an explicit semver; includes non-interactive release execution, GitHub Actions monitoring with gh, release note cleanup, and publishing the GitHub Release.

tomsawyerhu Bundle

Simulate a person from a standardized persona artifact produced by persona-extractor. Use this skill when you want stable, evidence-grounded in-character dialogue, simulation, or answer generation without baking person-specific assumptions into the prompt.

shader-slang Skill

Incrementally refresh the LLM-generated design docs under docs/generated/design/. Use when the user asks to update, refresh, or regenerate the generated design documentation after source changes. Runs the three-stage operator loop defined in docs/generated/design/_meta/regenerate.md - (1) regenerate stale docs (Claude), (2) direct the user to run the non-Claude review, (3) apply remediation (Claude) - with explicit confirmation before stages 1 and 3.

cloudflare Skill

Take a cloudflare/agents GitHub issue plus any repro findings and one-shot a fix PR — branch, change, test, push, and open the PR linked to the issue.

cloudflare Skill

Reproduce a cloudflare/agents GitHub issue by scaffolding a minimal Agents/Worker project and deploying it to a temporary Cloudflare account, then report findings back on the issue.

cloudflare Skill

Produce a focused test plan for a change. Use when the user asks how to test a feature, what cases to cover, or for a QA checklist before shipping.

cloudflare Bundle

Draft short release notes from a list of changes. Use when the user asks for changelogs, release notes, or a concise product update.

deepgram Skill

Use when writing or reviewing JavaScript/TypeScript in this repo that builds an interactive voice agent via `agent.deepgram.com/v1/agent/converse`. Covers `client.agent.v1.createConnection()` / `connect()`, `sendSettings`, `sendMedia`, runtime updates, event handling, and function-call responses. Use `deepgram-js-text-to-speech` for one-way synthesis, `deepgram-js-speech-to-text` or `deepgram-js-conversational-stt` for transcription only, and `deepgram-js-management-api` for project/model admin rather than live agent runtime. Triggers include "voice agent", "agent converse", "full duplex", "barge-in", "function calling", and "agent.v1".

deepgram Skill

Use when writing or reviewing JavaScript/TypeScript in this repo that calls Deepgram Speech-to-Text v1 (`/v1/listen`) for prerecorded or live audio transcription. Covers `client.listen.v1.media.transcribeUrl` / `transcribeFile` (REST) plus `client.listen.v1.createConnection()` / `connect()` (WebSocket). Use `deepgram-js-audio-intelligence` for summarize/sentiment/topics/diarize overlays, `deepgram-js-conversational-stt` for Flux turn-taking on `/v2/listen`, and `deepgram-js-voice-agent` for full-duplex assistants. Triggers include "transcribe", "speech to text", "STT", "listen.v1", "nova-3", "live transcription", and "websocket transcription".

deepgram Skill

Use when writing or reviewing JavaScript/TypeScript in this repo that calls Deepgram Text-to-Speech v1 (`/v1/speak`) for audio synthesis. Covers one-shot REST via `client.speak.v1.audio.generate` and streaming WebSocket via `client.speak.v1.createConnection()` / `connect()`. Use `deepgram-js-voice-agent` when you need full-duplex STT + LLM + TTS instead of one-way synthesis. Triggers include "TTS", "text to speech", "speak", "aura", "streaming TTS", and "speak.v1".

deepgram Skill

Use when writing or reviewing JavaScript/TypeScript in this repo that calls Deepgram Conversational STT v2 / Flux (`/v2/listen`) for turn-aware streaming transcription. Covers `client.listen.v2.createConnection()` / `connect()`, Flux models, and turn events like `TurnInfo`. Use `deepgram-js-speech-to-text` for standard v1 ASR and `deepgram-js-voice-agent` for full-duplex assistants. Triggers include "flux", "v2 listen", "conversational STT", "turn detection", "end of turn", "EOT", and "listen.v2".

guinacio Bundle

Generates professional AI images using Google Gemini. ALWAYS invoke this skill when building websites, landing pages, slide decks, presentations, or any task needing visual content. Invoke IMMEDIATELY when you detect image needs - don't wait for the user to ask. This skill handles prompt optimization and aspect ratio selection.

guinacio Bundle

Turn a generated or photographed image of a single object into a 3D mesh with the Tripo API, and download the GLB. Use when the request involves image-to-3D, photogrammetry-style reconstruction, turning a product shot or an isolated garment, prop or piece of hardware into a model for Blender, Unity or Unreal, or when Tripo, GLB output or 3D reconstruction credits come up. Not for generating the source image itself — the image-generation and character-reference-sheets skills cover that.

scotthuang Bundle

结构化任务规划与分步执行 V2（异步子代理架构）。支持将每个步骤通过 session_spawn 创建子代理异步执行，主会话保持非阻塞。功能包括：步骤分解、子代理执行、子代理检验（LLM判断）、状态跟踪、Heartbeat 监控、任务中断。触发词：/stp、任务规划、步骤执行。

chrbailey Skill

Analyze behavioral drift patterns across agents. Use when reviewing agent health, investigating anomalies, or during periodic governance reviews.

mistralai Skill

Authoritative reference for Mistral Vibe — the CLI agent you (the model) are running inside. LOAD when the user: - asks anything about Vibe itself, even by indirect name ("this CLI", "this tool", "you"); - wants to change, inspect, or reset their setup; - asks why the agent did or did not act; - asks how to make the CLI do X, where X lives, or what a flag/command/setting does; - asks any meta question about your own behavior; - is unsure whether a command, flag, env var, or file is in scope — this skill is the source of truth. SCOPE: config under `~/.vibe/` and project-local `.vibe/`; `VIBE_*` and `LOG_*` env vars; models and providers; agents and subagents; skills; tools and their permission model; every slash command and CLI flag; hooks; MCP servers; connectors; plugins; trusted folders; `@`-file mentions; logs; themes; voice.

mistralai Skill

Create a Vibe plugin package — the Agent Plugins 1.0 format with optional Vibe extensions for skills, MCP servers, hooks, knowledge, agents, libraries, and connectors. Use when the user wants to create, scaffold, or author a plugin for Mistral Vibe.

agentworkforce Skill

Use when writing agent-relay workflows that must fully validate features end-to-end before merging. Covers the 80-to-100 pattern - going beyond "code compiles" to "feature works, tested E2E locally." Includes PGlite for in-memory Postgres testing, mock sandbox patterns, test-fix-rerun loops, verify gates after every edit, and the full lifecycle from implementation through passing tests to commit.

agentworkforce Skill

Use when building multi-agent workflows with the relay broker-sdk - covers the WorkflowBuilder API, DAG step dependencies, agent definitions, step output chaining via {{steps.X.output}}, verification gates, evidence-based completion, owner decisions, dedicated channels, dynamic channel management (subscribe/unsubscribe/mute/unmute), swarm patterns, error handling, event listeners, step sizing rules, authoring best practices, and the lead+workers team pattern for complex steps