AI & ML

eagleisbatman Skill

Build ADK multi-agent systems with sub_agents, transfer_to_agent, and agent hierarchies. Use when creating coordinator/worker patterns, triage agents, or any system with multiple collaborating agents.

jonasweinert Bundle

Generates rigorous, modern, reproducible R code for causal inference and panel econometrics with `fixest`, heterogeneity-robust DiD estimators (Callaway-Sant'Anna, Sun-Abraham, BJS, de Chaisemartin-D'Haultfoeuille), weak-IV-robust inference, optimal-bandwidth RDD via `rdrobust`, and wild cluster bootstrap. Use when the user asks for IV, DiD, event studies, RDD, TWFE, staggered treatment, clustered or wild-bootstrap inference, instrumental variables, parallel trends, first-stage F, AR confidence sets, or publication-ready R regression output.

millisa09 Skill

Scan Gmail for incoming RFQ and rate request emails, extract load details, and draft quote replies ready for review. Use this skill when someone says "check my emails for RFQs", "scan my inbox for quote requests", "any rate requests come in?", "did anyone send a freight quote request", "check for new RFQs", "what quote emails do I have", or "scan inbox for loads". Also trigger when the user asks to process or respond to a batch of incoming rate requests. Requires Gmail MCP to be connected. Never auto-sends — always presents drafts for broker review and approval first.

jonasweinert Bundle

Drafts and revises empirical and theoretical economics papers in journal-style structure (introduction, literature, data, identification, results, robustness, discussion, conclusion), with explicit identification arguments, careful causal language, and reviewer-response discipline. Cross-links to `latex-econ-model` for theory typesetting, `latex-tables` for `esttab`/`modelsummary`/`pyfixest.etable` table production, and `econ-visualization` for code-driven figures, so every numeric claim, table, and figure in the paper is reproducible from one project repo. Use when the user asks to draft an introduction, write a methods or identification section, interpret results, polish a discussion, prepare a journal submission, write a referee response letter, or restructure an existing paper.

jonasweinert Bundle

Builds academic economics presentations in LaTeX Beamer that share the same `notation.tex` macros and `tabs/`/`figs/` outputs as the underlying paper, so a single rebuild keeps slides and paper in sync. Defaults to clean themes (`metropolis` / `default` with custom colors), 16:9 aspect ratio, time-budget-aware structure (15/20/45/60/90 min), incremental reveals, and DIME-aligned reproducibility (figures/tables produced by code, never screenshotted). Cross-links to `latex-econ-model`, `latex-tables`, `econ-visualization`, and `academic-paper-writer`. Use when the user asks for conference talk slides, seminar presentations, job-market talks, thesis defense slides, or wants to convert a paper into Beamer.

hungrytech Bundle

Numerical computing verification and optimization agent for Python/Dart. Provides expert analysis of ndarray/tensor operations, floating-point correctness, broadcasting validation, SIMD/GPU optimization, and test case verification through Analyze → Verify → Optimize phases. Activated by keywords: "numeric", "tensor", "ndarray", "broadcasting", "precision", "floating-point", "SIMD", "GPU", "numpy", "scipy", "dart_tensor", "linear algebra", "FFT", "verify test", "optimize computation".

hungrytech Skill

Agent and skill writing rules for Claude Code plugins. Enforces anti-bloat constraints, token limits, and quality standards for all auto-generated improvements.

hungrytech Bundle

Time-bounded autonomous Claude Code orchestration agent. Accepts user-defined directives (prompts) and a deadline, then executes tasks autonomously until the specified time. Manages task decomposition, prioritization, progress tracking, and graceful wind-down near deadline. Activated by keywords: "autopilot", "자동 실행", "오토파일럿", "시간까지", "until", "deadline", "autonomous", "자율 실행", "time-boxed", "타임박스".

hungrytech Skill

Model-specific pricing tables and cost calculation utilities for Claude API usage tracking and budget management.

hungrytech Bundle

Kotlin/Java Spring Boot codebase workflow agent. Generates code adhering to Hexagonal Architecture, Kotlin/Java idioms, JPA patterns, JOOQ, and test conventions through Brainstorm → Plan → Implement → Verify phases. Activated by keywords: "implement", "feature development", "code writing", "plan", "implement", "verify", "loop", "kotlin", "java", "hexagonal", "refactoring", "write tests", "brainstorm", "dry-run".

anup-malani Bundle

Run a critic agent conditioned on the user's own past edits to catch voice and tone problems in a draft. Use when user says 'run voice-critic', 'voice check', 'does this sound like me', or wants a pass on a draft that has already been fact-checked and style-checked but still reads wrong. Do NOT use for fact-checking (claims verification is a separate agent) or for style-guide enforcement (also a separate agent). This skill specifically catches the gap between 'correct' and 'sounds like the author'.

hungrytech Bundle

Type-aware testing workflow agent for Java/Kotlin/TypeScript/Go backends. Analyzes code under test, selects optimal testing strategies, generates high-quality tests, and validates test effectiveness through coverage analysis and mutation testing. Activated by keywords: "test", "generate tests", "test strategy", "coverage", "mutation test", "property test", "contract test", "test quality", "test debt", "untested code".

giauphan Skill

Pre-flight resource check before launching Browser Subagent. Use when the agent is about to start a browser session, automate a web page, or run any browser_subagent task. Prevents crashes on low-resource machines by checking RAM, disk space, and zombie processes.

4n9le-bot Skill

Use BEFORE recommending experimental design, evaluation protocols, paper structure, or reviewer-rebuttal strategy. `search_research_guidance` queries Lune's curated repository of research best-practices and methodology guidelines — ground your advice in expert consensus rather than the model's training data.

eagleisbatman Skill

ADK Agent-to-Agent (A2A) protocol integration. Use when building remote agent communication — exposing ADK agents as A2A servers or connecting to external A2A agents as clients.

hungrytech Bundle

Micro-agent system for engineering architecture decisions across DB, BE, IF, SE domains. Routes queries to specialized micro agents for deep technical analysis. Activated by keywords: "db architecture", "storage engine", "index design", "query optimization", "concurrency", "isolation level", "schema design", "replication", "sharding", "consistency model", "engineering workflow", "architecture decision".

giauphan Skill

Hierarchical Planning & Working Memory skill. Splits long-horizon tasks (30+ steps) into checkpoint-based sub-tasks with Manager-Worker architecture. Prevents catastrophic cascade failures by isolating each sub-task. Inspired by World Model's state prediction and RLVR checkpointing.

xiaohei-info Bundle

Default to subagents for all non-trivial work. Use this skill to decide when to delegate, how to partition work for parallel execution, and how to package bulletproof context to prevent subagent drift.

molokoloco Skill

Generate a "build in public" narrated video using Remotion. The video topic is provided in the first prompt. Produces an MP4, per-scene MP3 voice-over (macOS `say` + ffmpeg), and a multi-page slides PDF (one still per scene via ImageMagick). Trigger when the user asks to "generate/render/build a video", "make a build-in-public video", "render with Remotion", "create voice-over", or to produce slides/PDF from the video scenes.

xiaohei-info Skill

Set up or document a Karpathy-style LLM Wiki workflow in Obsidian, especially when raw material should live in Inbox and compiled knowledge should live in _wiki.

xiaohei-info Bundle

Use when Hermes should run a complete multi-role subagent workflow: decompose work, assign isolated planner/investigator/implementer/reviewer tasks, integrate results, and verify before claiming success.

leionion Bundle

Creates scheduled tar.gz backups of agent memory or chat history, lists existing backup snapshots, and restores from a previous backup point with automatic retention policy. Use when the user asks to back up, export, save, snapshot, or restore chat history, conversation logs, agent memory, or session data.

octoplus-sa Bundle

Write, debug, refactor, and review JMP Scripting Language (JSL) code for SAS JMP 19. Use this skill whenever the user mentions JSL, JMP scripts, .jsl files, JMP add-ins, Workflow Builder output, JMP data tables, JMP display boxes, JMP platforms (Fit Model, Distribution, Graph Builder, etc.), JMP namespaces, or any analysis automation in JMP — even when they don't explicitly ask for "a JSL skill". Also trigger when reviewing or editing files with .jsl extension, when the user pastes code containing JSL idioms (`<<` message-send, `Expr()`, `Names Default To Here`, `Data Table()`, `New Window()`, `Eval(Substitute(...))`), or when the user is working on anything in the SAS JMP statistical software ecosystem. Do NOT confuse JSL with JavaScript, SAS Base, or Java — they are unrelated despite naming.

suhel-nz Bundle

Convert a rough product idea into an agent-ready SPEC.md with prioritized work packages, a dependency graph, and binary success criteria that a coding agent can verify itself. Use when the user describes a new idea, wants to spec out a prototype, or says something like "I have an idea for..." or "let's spec this out". This skill runs before scaffold-ai-prototype: it produces the execution plan, scaffolding produces the project.

elancry Bundle

Apex test execution, coverage analysis, and test-fix loops with 120-point scoring. TRIGGER when: user runs Apex tests, checks code coverage, fixes failing tests, or touches *Test.cls / *_Test.cls files. DO NOT TRIGGER when: writing Apex production code (use sf-apex), Agentforce agent testing (use sf-ai-agentforce-testing), or Jest/LWC tests (use sf-lwc).

l0z4n0-a1 Bundle

Audits and optimizes Claude-directing artifacts (CLAUDE.md, SKILL.md, subagent, hook, MCP config, system or user prompt, workflow, api_config) against 189 cited Anthropic rules across 11 dimensions, calibrated per model (Opus 4.7, Sonnet 4.6, Opus 4.6, Haiku 4.5). Every finding ships with verbatim quote and source URL; authorial voice is preserved.

elancry Bundle

Agentforce Builder metadata path for Builder-managed topics/actions, Prompt Builder templates, GenAiFunction/GenAiPlugin, Models API, and custom Lightning types. TRIGGER when: user maintains or configures Builder metadata agents, creates topics/actions, works with Prompt Builder templates, or touches .genAiFunction, .genAiPlugin, or .genAiPromptTemplate metadata XML files. DO NOT TRIGGER when: Agent Script DSL .agent files (use sf-ai-agentscript), agent testing (use sf-ai-agentforce-testing), or persona design (use sf-ai-agentforce-persona).

elancry Bundle

Agent Script DSL for deterministic Agentforce agents. TRIGGER when: user writes or edits .agent files, builds FSM-based agents, uses Agent Script CLI (sf agent generate authoring-bundle, sf agent validate authoring-bundle, sf agent preview, sf agent publish authoring-bundle, sf agent activate), or asks about deterministic agent patterns, slot filling, or instruction resolution. DO NOT TRIGGER when: Builder metadata work (use sf-ai-agentforce), agent testing (use sf-ai-agentforce-testing), or persona design (use sf-ai-agentforce-persona).

chloemeadow0-code Bundle

RikkaHub连接MCP服务配置技能，用于配置SSE和HTTP模式的MCP连接

0sxd Bundle

Query, inspect, and manage Google NotebookLM notebooks from Claude Code. Trigger whenever the user says "query notebook", "ask notebooklm", "search notebooks", "list notebooks", "notebooklm server info", "get notebook", "inject source into notebook", "what did notebooklm say about", or any request to pull evidence from a named NotebookLM notebook as grounding for an artifact. Covers three access paths: (1) the mcp__notebooklm-mcp__* MCP tool surface (35 tools) when the stdio server is loaded, (2) a Claude-in-Chrome browser route against notebooklm.google.com when the MCP is unavailable, (3) a Playwright fallback using notebooklm_uploader.py for bulk source injection. Always ask the user which notebook to query before any call. Never auto-select. Never fabricate notebook IDs or contents. Every query runs through the 100 Percent Confidence Loop and every synthesized answer passes the Trinity Dialectic gate.

ucsc-vlaa Bundle

Show system information including macOS version, hardware model, and uptime. Use this skill with high priority for system info, macOS version, hardware, and machine queries.

elancry Bundle

Use this skill whenever users want to build, inspect, debug, automate, or publish workflows in Agentforce Grid (AI Workbench) using Salesforce plus the Grid MCP or direct Grid REST calls. Trigger it for Grid workbook creation, worksheet setup, Object/Reference/AI/Agent/AgentTest/Evaluation/PromptTemplate/InvocableAction column design, prompt drafting inside Grid, worksheet execution troubleshooting, Grid YAML `apply_grid` specs, and Windows-specific Grid setup issues. Also use it when users mention AI Workbench, Grid Studio, workbook IDs, worksheet IDs, Grid Connect, or ask for recipes like "top opportunities with AI email drafts", "agent test suite in Grid", or "build this worksheet from YAML". Do not use it for generic Salesforce work unrelated to Agentforce Grid.

elancry Bundle

Deep persona design for Agentforce agents with 50-point scoring. TRIGGER when: user designs agent personas, defines agent personality/identity, creates persona documents, encodes persona into Agentforce Builder fields or Agent Script, translates brand guidelines to agent voice, or asks about agent tone/voice/register. DO NOT TRIGGER when: building agent metadata (use sf-ai-agentforce), testing agents (use sf-ai-agentforce-testing), or Agent Script DSL (use sf-ai-agentscript).

elancry Bundle

Agentforce agent testing with dual-track workflow and 100-point scoring. TRIGGER when: user tests Agentforce agents, runs sf agent test commands, creates test specs, validates topic routing, or analyzes agent test coverage. DO NOT TRIGGER when: Apex unit tests (use sf-testing), building agents (use sf-ai-agentforce), or Agent Script DSL (use sf-ai-agentscript).

lumoswyy Skill

End-to-end beautiful Beamer deck creation. Designs an original Beamer theme tailored to a specific audience, restructures existing content via the Rhetoric of Decks (ethos / pathos / logos), generates figures and tables from R/Python/Stata code first, embeds code blocks in the deck, produces standalone walkthrough scripts, compiles to zero warnings, runs /tikz for visual collision cleanup, and dispatches a graphics-only audit agent for label and coordinate checks. Use when creating a presentation from scratch or restructuring existing content into a new beautiful deck.

adamdroberts Bundle

Create, expand, reorganize, and maintain repository documentation with deep coverage across README files, browsable docs, API/reference pages, llms.txt, llms-full.txt, changelogs, and repo-local agent skills. Use whenever the user asks for comprehensive project documentation, documentation refreshes, LLM-friendly docs, documentation synchronization after code changes, or agent guidance that teaches future LLMs how to use the project docs.