Top Agent Skills

Use to prove that a support process, policy or script change actually took effect — when, for whom, and whether behaviour followed — rather than that it was approved. Trigger for "prove we made that change", "when did this policy actually change", "did the team adopt the new process", evidencing remediation, "we told everyone" claims, or a change that was signed off and never landed.

Use to set up or fix quality assurance for voice/phone support, including transcript-based and AI-graded call QA. Trigger for "QA our calls", "score phone conversations", call transcription accuracy for QA, diarisation or speaker attribution problems, word error rate, grading calls from transcripts, or adapting a chat/email QA scorecard to voice.

Use to review how an outsourcer, BPO or vendor handles your customer data and meets your conduct obligations, using evidence from the work rather than from their questionnaire answers. Trigger for "review our BPO's data handling", "vendor risk assessment for our outsourcer", "are our partners compliant", third-party oversight, outsourcing due diligence, or a supplier audit of a support vendor.

Use to decide how to split a growing support organisation — by product, segment, channel, or complexity — and when pods beat shared queues, accounting for coordination cost. Trigger for "team topology", "how should we split the team", "organise support by product or queue", "when to create pods", "support team structure", growing headcount, or coordination overhead between sites.

Use to scan support conversations for a harm nobody has categorised yet — a new failure mode, a scam pattern, or a product change hurting a group of customers before it shows in any metric. Trigger for "is anything new going wrong", "scan for emerging issues", "early warning from support", horizon scanning, a new scam pattern, or preparing for a question about what you might be missing.

Use to audit arrears, collections and financial-difficulty conversations for fair treatment — forbearance offered, pressure absent, vulnerability recognised. Trigger for "audit our collections calls", "are we treating customers in arrears fairly", "did we offer forbearance", financial difficulty handling, payment plan conversations, or a complaint about collections conduct.

Use to assign ownership, set review cadence, and deprecate support content without breaking self-service or AI grounding. Trigger for "who owns this article", "our KB is out of date", orphan articles, annual review calendar nobody follows, deprecating help content, knowledge governance, or preventing stale articles from quietly harming customers.

Use to clean up sprawling helpdesk tags, categories and custom fields so reporting built on them means something. Trigger for "our tags are a mess", "consolidate our ticket categories", "why don't our tag reports add up", duplicate or near-duplicate tags, tags nobody uses, agents tagging inconsistently, or before building reporting on top of tags.

Configure, troubleshoot, and use Hermes image generation — all 5 built-in providers (FAL, OpenAI, OpenAI-Codex, xAI, Krea), model catalogs, env vars, and .env location gotchas.

Design data systems by understanding storage engines, replication, partitioning, transactions, and consistency models. Use when the user mentions "database choice", "replication lag", "partitioning strategy", "consistency vs availability", "stream processing", "ACID transactions", "eventual consistency", or "LSM tree vs B-tree". Also trigger when choosing between SQL and NoSQL, designing data pipelines, or debugging distributed system consistency issues. Covers data models, batch/stream processing, and distributed consensus. For system design, see system-design. For resilience, see release-it.

Embedding database for RAG and semantic search.

Poll RSS, JSON APIs, and GitHub with watermark dedup.

Expert skill for building LLM applications with the LlamaIndex framework — RAG pipelines, multi-agent orchestration, event-driven workflows, knowledge graph construction, production deployment, and evaluation. Use when working with LlamaIndex or comparing RAG and agent orchestration frameworks.

A virtual Kanban expert who can diagnose flow problems, design board configurations, set up multi-portfolio operating models, calibrate WIP limits, establish service level expectations, and guide Scrum-to-Kanban transitions. Load this when your team is struggling with throughput, cycle times are unpredictable, multiple stakeholders compete for the same engineers, or you're wondering if Kanban is right for you.

Apply named refactoring transformations to improve code structure without changing behavior. Use when the user mentions "refactor this", "code smells", "extract method", "replace conditional", "technical debt", "move method", "inline variable", or "decompose conditional". Also trigger when cleaning up legacy code, preparing code for new features by restructuring, or identifying which transformation to apply to a specific code smell. Covers smell-driven refactoring, safe transformation sequences, and testing guards. For code quality foundations, see clean-code. For managing complexity, see software-design-philosophy.

DSPy: declarative LM programs, auto-optimize prompts, RAG.

Monitor the local Mac and Hermes runtime with real scripts, including context-aware anomaly detection and a compact runtime status report.

Search and download GIFs from Tenor using curl. No dependencies beyond curl and jq. Useful for finding reaction GIFs, creating visual content, and sending GIFs in chat.

Pixel art w/ era palettes (NES, Game Boy, PICO-8).

Build complex AI systems with declarative programming, optimize prompts automatically, create modular RAG systems and agents with DSPy - Stanford NLP's framework for systematic LM programming

Query Polymarket prediction market data — search markets, get prices, orderbooks, and price history. Read-only via public REST APIs, no API key needed.

Create, manage, triage, and close GitHub issues. Search existing issues, add labels, assign people, and link to PRs. Works with gh CLI or falls back to git + GitHub REST API via curl.

Monitor blogs and RSS/Atom feeds for updates using the blogwatcher-cli tool. Add blogs, scan for new articles, track read status, and filter by category.

Airtable REST API via curl. Records CRUD, filters, upserts.

PyTorch library for audio generation including text-to-music (MusicGen) and text-to-sound (AudioGen). Use when you need to generate music from text descriptions, create sound effects, or perform melody-conditioned music generation.

Full pull request lifecycle — create branches, commit changes, open PRs, monitor CI status, auto-fix failures, and merge. Works with gh CLI or falls back to git + GitHub REST API via curl.

Generate project ideas through creative constraints. Use when the user says 'I want to build something', 'give me a project idea', 'I'm bored', 'what should I make', 'inspire me', or any variant of 'I have tools but no direction'. Works for code, art, hardware, writing, tools, and anything that can be made.

Generate professional dark-themed system architecture diagrams as standalone HTML/SVG files. Self-contained output with no external dependencies. Based on Cocoon AI's architecture-diagram-generator (MIT).

Delegate coding tasks to Claude Code (Anthropic's CLI agent). Use for building features, refactoring, PR reviews, and iterative coding. Requires the claude CLI installed.

Track ML experiments with automatic logging, visualize training in real-time, optimize hyperparameters with sweeps, and manage model registry with W&B - collaborative MLOps platform

Use when encountering any bug, test failure, or unexpected behavior. 4-phase root cause investigation — NO fixes without understanding the problem first.

Pattern for deprecating external dependencies while maintaining backward compatibility. Trigger: removing Java/FIJI/MATLAB, deprecating optional deps

Audit SQLite persistence layer for unused tables and broken integrations. Trigger when: (1) checking database usage, (2) cleaning up schema, (3) finding missing methods.

SLURM EDF job must use marker names from CHANNELNAMES.txt for output files. Trigger: EDF output naming, SLURM job producing CH#_edf.tif instead of marker names, downstream _find_edf_file failures

Pattern for creating new Colab notebooks. Trigger when: (1) Creating a new notebook for experiments, (2) Adding notebook-based functionality, (3) Agent or validation notebooks, (4) Any notebook that uses GPU training infrastructure.

Build predictive models for sports and event outcomes using statistical methods, ELO ratings, regression, Monte Carlo simulation, and machine learning. Use when creating power rankings, projecting game outcomes, estimating win probabilities, or building a quantitative edge. Also trigger for 'prediction model', 'ELO rating', 'power rankings', 'win probability', 'Monte Carlo', 'regression model', 'expected goals', or 'predictive analytics'.

Analyze betting market sentiment including line movement, sharp vs public money, steam moves, reverse line movement, and market signals. Use when reading line movement to determine where the smart money is going. Also trigger for 'line movement', 'sharp money', 'public money', 'steam move', 'reverse line movement', 'RLM', 'where the money is', or 'wiseguy action'.

Compact the current session into a single detailed handoff message that can be pasted into a fresh agent run. Use when switching context, ending a session, or avoiding context-window loss.

Draft a detailed Non-Disclosure Agreement between two parties covering information types, jurisdiction, and clauses needing legal review. Use when creating confidentiality agreements or preparing an NDA for a partnership.

Use Orca orchestration for structured multi-agent coordination: threaded messages, blocking ask/reply flows, task dispatch, worker_done/escalation waits, task DAGs, decision gates, coordinator loops, or decomposing work across agents. Use `orca-cli` instead for full ownership handoffs, including requests phrased as "hand off", "handoff", "handover", "give this to another agent", or "another worktree" when the user did not explicitly ask to supervise, monitor, wait for results, or coordinate a DAG. Use `orca-cli` for ordinary terminal control, lightweight terminal prompts, shell commands, Orca worktree management, reading or waiting on terminals, and automation of the browser embedded inside Orca. Use Computer Use for browser windows, webviews, Orca app UI, or desktop UI outside Orca's embedded browser.

Generate SQL queries from natural language descriptions. Supports BigQuery, PostgreSQL, MySQL, and other dialects. Reads database schemas from uploaded diagrams or documentation. Use when writing SQL, building data reports, exploring databases, or translating business questions into queries.

Delegate coding to xAI Grok Build CLI (features, PRs).

Create refined user personas from research data — 3 personas with JTBD, pains, gains, and unexpected insights. Use when building personas from survey data, creating user profiles from research, or segmenting users for product decisions.

Generate a Startup Canvas combining Product Strategy (9 sections) and Business Model (costs + revenue) for a new product. An alternative to BMC and Lean Canvas that separates strategy from business model. Use when launching a new product or evaluating a startup concept.

中英双语学术降 AIGC / bilingual academic de-AIGC skill. Removes AI-generated writing signatures from empirical papers in economics, management, and the social sciences — in both English and Chinese. Covers Turnitin AI, GPTZero, Originality.ai on the English side and 知网 AMLC, 万方, 维普 on the Chinese side. Uses a six-step loop (intake → audit → claim-evidence check → differentiated rewrite → five-dimension self-score → cold-reader recheck) with two pattern libraries (22 English + 17 Chinese patterns), section-by-section strategies for empirical papers, and hard protections that keep every number, coefficient, and citation intact.

Remove AI writing patterns from prose. Use this skill when writing, drafting, editing, reviewing, or revising any text to eliminate predictable AI tells, slop, and formulaic patterns. Trigger this skill whenever the user asks to "deslop", "de-AI", "make it sound human," "remove AI patterns," "remove AI tropes," "clean up AI writing," fix "slop," "deslop" text, or review prose for authenticity. Also use when the user asks you to write or draft anything and wants it to sound natural rather than AI-generated. Common use cases include scientific writing (manuscripts, abstracts, cover letters, grant narratives, discussion sections, peer review responses), blog posts, newsletters, memos, reports, and any other substantial prose.

Workflow 4: Submission rebuttal pipeline. Parses external reviews, enforces coverage and grounding, drafts a safe text-only rebuttal under venue limits, and manages follow-up rounds. Use when user says "rebuttal", "reply to reviewers", "ICML rebuttal", "OpenReview response", or wants to answer external reviews safely.

Use when a research task needs reproducible Kaggle discovery, metadata inspection, bounded public-data downloads, competition or kernel discovery, model discovery, or an explicitly approved Kaggle write/delete operation through the official CLI.

Classical end-to-end empirical analysis workflow in the modern tidyverse + econometrics R ecosystem — dplyr + tidyr + haven + fixest + sandwich + lmtest + clubSandwich + AER + ivreg + did + bacondecomp + HonestDiD + eventstudyr + rdrobust + rddensity + Synth + gsynth + synthdid + MatchIt + WeightIt + cobalt + ebal + grf + DoubleML + mediation + marginaleffects + modelsummary + kableExtra + gt + ggplot2 + ggpubr + cowplot + binsreg. **Defaults to economics empirical-paper style** (AER / QJE / AEJ) — every run produces a publication-ready output set with a multi-column regression table (M1→M6 progressive controls/FE) as the centerpiece, plus Table 1 (descriptives), mechanism / heterogeneity / robustness tables, and event-study + coefficient + trend figures. Covers the full 8-step R pipeline an applied economist runs on every paper — (1) data import & cleaning (read_dta/read_csv, naniar, janitor, validate-merges), (2) variable construction (mutate/across/winsorize/group_by + lag/lead with dplyr), (3) descriptive

Install, configure, update, or verify mcp-stata across Claude Code, Codex, Gemini CLI, Cursor, Windsurf, and VS Code. Activate when users ask to set up the Stata toolkit or troubleshoot the installation.

Brainstorm a longlist of benefits and costs for a project. Outputs two clean tables (benefits and costs), each with materiality (H/M/L), a cash flow tag (cash in / cash out / non-cash), a quantification method, and a monetisation method. Supports HMT Green Book, EU Better Regulation Guidelines, World Bank, Asian Development Bank, and the Victorian Treasury High Value High Risk (HVHR) framework.

STATA code pattern library for empirical archival accounting research. Provides tested syntax from 126 peer-reviewed JAR (Journal of Accounting Research) replication files (2017-2025). Use when the user asks procedural questions like "How do I implement [method]?" or "Show me code for [technique]" — including: entropy balancing, propensity score matching (PSM), difference-in-differences (DiD), regression discontinuity (RDD), instrumental variables (IV), event studies (CAR/BHAR), survival analysis, Fama-MacBeth regressions, bootstrap, quantile regression, reghdfe/xtreg/areg, clustering standard errors, fixed effects, esttab/outreg2 table formatting, winsorization, leads/lags. Users can specify their variables (e.g., treatment, outcomes, controls) and receive adapted syntax. NOTE: This skill provides code patterns from published papers, not research design advice.

Verify academic citations against CrossRef, Semantic Scholar, and OpenAlex. Detects AI-hallucinated references, chimeric citations, and suspicious patterns.

Classical end-to-end empirical analysis workflow in the traditional Stata ecosystem — native Stata + reghdfe + ivreg2 + csdid + did_imputation + eventstudyinteract + sdid + rdrobust + rddensity + synth + synth_runner + psmatch2 + teffects + ebalance + coefplot + esttab + asdoc + binscatter. **Defaults to economics empirical-paper style** (AER / QJE / AEJ) — every run produces a publication-ready output set with a multi-column regression table (M1→M6 progressive controls/FE) as the centerpiece, plus Table 1 (descriptives), mechanism / heterogeneity / robustness tables, and event-study + coefficient + trend figures. Covers the full 8-step Stata pipeline an applied economist runs on every paper — (1) data import & cleaning (use/import, destring, misstable, duplicates, merge assert), (2) variable construction (gen/egen/winsor2/xtile/xtset with L./F./D.), (3) descriptive statistics & Table 1 (tabstat/balancetable/asdoc), (4) classical diagnostic tests (sktest/swilk/hettest/imtest/xtserial/xttest3/vif/dfuller/kpss/

Profile a target (script, process, GPU, memory, interconnect) using external tools and code instrumentation. Produces structured performance reports with actionable recommendations. Use when user says "profile", "benchmark", "bottleneck", or wants performance analysis.

Proofread and verify academic papers in LaTeX. Runs 19 sequential checks covering titles, consistency, citations, formatting, theoretical tension in motivation, concise results reporting, cross-section repetition, em-dash usage, and auxiliary-text-to-footnote conversion. Trigger when user says "check paper" / "proofread" / "paper-checker" / "校对" / "核查论文".

Turn a vague research direction into a problem-anchored, elegant, frontier-aware, implementation-oriented method plan via iterative GPT-5.4 review. Use when the user says "refine my approach", "帮我细化方案", "decompose this problem", "打磨idea", "refine research plan", "细化研究方案", or wants a concrete research method that stays simple, focused, and top-venue ready instead of a vague or overbuilt idea.

Improve, modernize, and optimize existing Stata code for performance, portability, and maintainability. Use when legacy patterns such as preserve/restore, cd,

Generate professional slide deck images from academic papers and content. Creates comprehensive outlines with style instructions, auto-detects figures from PDFs, then generates individual slide images. Use when user asks to "create slides", "make a presentation", "generate deck", or "slide deck" for papers.

Audit datasets for structure, missingness, labeling, suspicious values, duplicate identifiers, and documentation readiness. Use when a researcher asks for data QA, codebook review, sanity checks, or pre-analysis cleanup guidance.