Skills
100,928 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Analyze experiment results and generate discussion paragraphs for academic papers. Two-phase workflow: identify measurable findings (Phase 1), confirm with user, then generate grounded discussion paragraphs (Phase 2). Accepts tables, statistics, or result descriptions. 实验分析与讨论段落生成。
Set up a structured project memory for any software project — TODO.md as priority/effort-tagged index of future work, backlog/ for resume-friendly research/design notes on P? items, and pitfalls/ as a symptom-grep-able knowledge base of past traps. Use when a user wants somewhere to record "maybe later" ideas, freeze troubleshooting state, capture trade-off analysis, or stop re-debugging the same problem.
Query PubChem (110M+ compounds) directly via the PUG-REST/JSON API with plain `requests` — no SDK install required. Search by name/CID/SMILES/InChIKey/formula, retrieve properties (MW, XLogP, TPSA, H-bond counts), do similarity/substructure searches with async ListKey polling, fetch synonyms, descriptions, assay summaries, and download SDF/PNG. For local cheminformatics use rdkit; for bioactivity-centric workflows use chembl-database-bioactivity.
Direct REST API access to PubMed. Advanced Boolean/MeSH queries, E-utilities API, batch processing, citation management. For Python workflows, prefer biopython (Bio.Entrez). Use this for direct HTTP/REST work or custom API implementations.
High-performance reinforcement learning framework optimized for speed and scale. Use when you need fast parallel training, vectorized environments, multi-agent systems, or integration with game environments (Atari, Procgen, NetHack). Achieves 2-10x speedups over standard implementations. For quick prototyping or standard algorithm implementations with extensive documentation, use stable-baselines3 instead.
Comprehensive healthcare AI toolkit for developing, testing, and deploying machine learning models with clinical data. This skill should be used when working with electronic health records (EHR), clinical prediction tasks (mortality, readmission, drug recommendation), medical coding systems (ICD, NDC, ATC), physiological signals (EEG, ECG), healthcare datasets (MIMIC-III/IV, eICU, OMOP), or implementing deep learning models for healthcare applications (RETAIN, SafeDrug, Transformer, GNN).
Python Materials Genomics library for structure analysis, thermodynamics, and electronic properties. Parse/create crystal structures (CIF, POSCAR), query Materials Project for DFT-computed properties, analyze phase and Pourbaix diagrams, compute XRD patterns, generate DFT inputs for VASP, Quantum ESPRESSO, CP2K. Alternatives: ASE (MD/geometry), AFLOW (high-throughput), OVITO (visualization).
Professional data visualization using Python (matplotlib, seaborn, plotly). Create publication-quality static charts, statistical visualizations, and interactive plots. Use when generating charts/graphs/plots from data, creating infographics with data components, or producing scientific/statistical visualizations. Supports PNG/SVG (static) and HTML (interactive) export.
Complete Python testing system. PROACTIVELY activate for: (1) pytest fundamentals, (2) Fixtures and scopes, (3) Parameterized tests, (4) Mocking with pytest-mock, (5) Async testing with pytest-asyncio, (6) Coverage configuration, (7) FastAPI testing with httpx, (8) Property-based testing with hypothesis, (9) Snapshot testing. Provides: pytest patterns, fixture examples, mock setup, coverage config. Ensures comprehensive test coverage with best practices.
World-class QA engineering - systematic testing, automation, and the mindset that finds bugs before users doUse when "QA, quality assurance, testing, test automation, e2e tests, integration tests, regression testing, test coverage, playwright, cypress, selenium, test suite, bug report, test strategy, flaky tests, testing, QA, automation, e2e, integration, regression, quality" mentioned.
Create, customize, or polish application-ready resumes from scratch for university students, new graduates, and internship seekers: supports a Grill Me-style question-and-answer interview to collect experience from zero and generate a PDF; reads job JDs and old resumes to build a requirement–evidence mapping and rewrite accordingly; optimizes the content structure of an existing resume or only polishes the layout without changing facts; and generates six different styles of ATS-friendly PDF resumes at once when the user explicitly requests it. Applies to requests such as "interview me to make a resume", "no resume, Q&A from scratch", "customize a resume based on a JD", "optimize/rearrange/polish an existing resume", and "generate multiple resume styles at once"; supports Chinese/English and computer science, software, AI, data, and other campus recruitment positions. Uses the A-C-R-E structure to review experience and adapts content density as sparse / balanced / dense. Fabricating experience or numbers is prohibited; by default it generates one most suitable single-column resume, and only generates in bulk when the user explicitly asks for a comparison. Do not use for academic application CVs, senior executive resumes, portfolio websites, submitting applications on someone's behalf, or pure mock interviews.
Build, simulate, transpile, and execute quantum circuits with Qiskit and IBM Quantum Runtime. Use for Qiskit 2.x circuits and operators, V2 Sampler or Estimator primitives, target-aware transpilation, local or noisy simulation, IBM QPU execution, Runtime sessions or batches, error mitigation, and Qiskit ecosystem packages.
Interpret factor/stock-selection logic from strategy descriptions, research report PDFs, or screenshots, and generate a QMT built-in daily-frequency factor backtest script based on the scripts/daily-factors-backtest.py framework. Use when the user mentions QMT built-in backtesting, factor stock-selection backtesting, cross-sectional factors, research report replication, handlebar backtesting, or after_init precomputed signals.
One-sentence financial onboarding (and offboarding) of an employee on Qonto. Reads a per-role policy defined once with the user (card caps, online-only, team), then on "Alex starts Monday as a developer" it invites the member, creates the team if new, and prepares a capped card request that conforms to the policy. Offboarding is the mirror image - freeze the cards, generate a recovery checklist, delete nothing. Use for explicit Qonto financial onboarding or offboarding requests such as "onboard Sam as an account exec" or "Sam is leaving Friday — freeze his Qonto cards".
Run ClickHouse `system.query_log` analysis via the internal Metabase API. Use when investigating slow queries, materialization candidates, per-team query performance, ClickHouse cost or memory issues, or any system.query_log question. Covers prod-us and prod-eu, SSO-gated cookie auth via `hogli`, and ready-to-run query patterns.
Invent and run exploratory browser test cases for a web app with quicke2e. Maps the app, reads its source for business rules, writes goal-based specs with deterministic assertions, runs them with a small decision model (hosted Jev or a local model), and reports findings. Use when the user says “quicke2e,” “explore my app,” “find UI bugs,” “write browser test cases,” or wants adversarial end-to-end checks without writing selectors.
Guides modern Rails 8 code architecture decisions and patterns. Use when deciding where to put code, choosing between patterns (service objects vs concerns vs query objects), designing feature architecture, refactoring for better organization, or when user mentions architecture, code organization, design patterns, or layered design. WHEN NOT: Implementing specific patterns (use specialist agents like service-agent or query-agent), writing tests, or debugging runtime errors.
The sole entry point for reverse engineering analysis. Workflow: environment probe (probe.sh) → ask preferences (analysis target/decompiler/depth/report/platform) → identify task (triage.md) → orchestrate calls to category gateways. Triggers: analysis, reverse engineering, reverse, RE, “help me look at this file,” “what does this sample do,” unpacking, finding keys, cracking, firmware analysis, malware analysis, “analyze this binary.”
Software reverse engineering core gateway (shared foundation). Orchestrates: initial survey → format parsing → decompilation → debugging/tracing → memory. Sub-skills: [[re-address-space]] [[re-triage]] [[re-format-pe]] [[re-format-elf]] [[re-format-macho]] [[re-imports]] [[re-ghidra]] [[re-ida]] [[re-radare2]] [[re-gdb]] [[re-x64dbg]] [[re-lldb]] [[re-tracing]] [[re-memdump]] [[re-windbg]] [[re-binaryninja]] [[re-emulation]] [[re-shellcode]] [[re-kernel]] [[re-game]] [[re-go]] [[re-rust]] [[re-plugin-dev]] [[re-hypervisor]] [[re-anti-cheat]], [[re-cpp-abi]], [[re-swift]], [[re-zig]], [[re-nim]], [[re-fp-runtime]], [[re-variant]], [[re-mips]], [[re-arm]], [[re-riscv]]. Triggers: static analysis, look at this program's logic, decompilation, reverse this binary, binary analysis.
Editing Russian-language texts using Lyudmila Sarycheva's method — from meaning to words, not the other way around. Use this skill whenever the user asks to edit, review, improve, shorten, or polish any Russian-language text — articles, posts, emails, landing copy, docs. Triggers: «отредачь», «отредактируй», «почисти текст», «сократи», «перепиши по-человечески», «убери канцелярит», «сделай проще», «причешите текст», «дай фидбек на текст», «разбери мой текст», «слишком водянисто», «звучит как робот». Also trigger when the user pastes a Russian draft and asks what's wrong with it or how to make it better. For writing text from scratch, see skill statya; for interface texts — ux-copy.
Safely identify and remove dead code with test verification at every step. Use when Codex needs to clean unused exports, files, dependencies, helpers, wrappers, re-exports, or duplicate code without risking regressions. Trigger for dead-code cleanup, unused dependency audits, stale utility removal, cleanup refactors, or requests to reduce code safely before or alongside maintenance work.
This skill should be used when bootstrapping, auditing, or reviewing the community-health and repository-cornerstone files in a code repository: README, LICENSE, CONTRIBUTING, CODE_OF_CONDUCT, SECURITY, SUPPORT, issue / PR templates, CODEOWNERS, FUNDING, CITATION, and REUSE / SPDX metadata. PROACTIVELY activate on "set up new repo docs", "community health files", "what should my README contain", "do I need a code of conduct", "audit repo bootstrap docs", "repository documentation cornerstones", "pick a license", "CONTRIBUTING.md", "SECURITY.md", "vulnerability reporting", "SUPPORT.md", "CODEOWNERS", "issue templates", "PR templates", "CITATION.cff", "REUSE", "SPDX headers", "Standard Readme", "Contributor Covenant". Provides: four-question diagnostic per cornerstone, Standard Readme structure, license routing (no picking), CONTRIBUTING / CODE_OF_CONDUCT / SECURITY / SUPPORT canons, template files, and REUSE 3.3 / SPDX routing.
Deep-research a topic end to end and produce a permanent, reusable knowledge asset: archive every primary source verbatim (gated by the user's privacy/retention posture), write one 1:1 summary per source, then synthesize a single self-contained compendium page. Depth is a dial (base synthesis → grounded primaries → books + counter-canon → saturation), each level an idempotent superset of the one below. Distinct from data-research (structured trackers) and perplexity-research (web deltas): this produces prose knowledge synthesis backed by an archived source corpus.
Research paper writing assistant with 30 Prompt templates covering the full paper-writing process. Use when users mention paper writing, polishing, translation, literature reviews, abstracts, introductions, methods sections, reviewer responses, grant applications, research plans, academic presentation slides, LaTeX editing, academic charts, experimental results analysis, simulated peer review, removing AI-like wording, literature comparisons, finding research gaps, or paper outlines. Do not use for non-academic writing, creative writing, blogs, marketing copy, cover letters, novels, or similar contexts.
Read-only multi-agent review of a GitHub Pull Request, with the synthesized report posted back as a PR comment so the author is notified. Use when the user wants to review a GitHub PR (github.com or GitHub Enterprise) and post a structured review back to the PR conversation; the PR is auto-detected from the checked-out branch when no locator is supplied. Activates the `review-anvil` engine in read-only mode.
Use this skill when a PR diff contains changes to dependency manifest files (package.json, .csproj, Cargo.toml, go.mod, requirements.txt, etc.) or when reviewing Renovate/Dependabot bot PRs. Evaluates new dependencies for AppSec approval process compliance, major version bump significance, lock file hygiene, and dependency removal completeness. Does NOT perform deep security or license analysis — that is handled by the bitwarden-security-engineer plugin's reviewing-dependencies skill.
Use for Roblox Open Cloud and external integration work: choosing API-key-based authentication for server-to-server automation, constructing and troubleshooting Open Cloud REST requests, checking scopes and rate limits, determining whether HttpService can call an endpoint, setting up webhook receivers, and using Roblox OpenAPI and reference JSON artifacts for tooling or client generation.
Provides comprehensive guidelines for resolving merge conflicts intelligently using git history and commit context. Use when tasks involve merge conflicts, rebasing, PR conflicts, or git conflict resolution. This skill analyzes commit messages, git blame, and code intent to make intelligent resolution decisions.
Use this skill whenever the user asks for a security analysis, vulnerability assessment, security audit, or any form of Security Assessment Report (SAR) over a codebase, infrastructure, API, database, or system. Triggers include: "audit my code", "find security issues", "run a security check", "generate a SAR", "check for vulnerabilities", "is this code secure", or any request that involves evaluating the security posture of a project, repository, service, or infrastructure. For a single snippet or a quick "is this secure?" question, answer inline and offer a full SAR instead of running it. Do NOT use for generic coding tasks, code reviews focused on quality rather than security, or performance optimization unless a security angle is explicitly present.
Detect insecure file upload vulnerabilities in a codebase using a three-phase approach: discovery (find all upload sites), batched verify (check extension bypass and related issues in parallel subagents, 3 sites each), and merge (consolidate batch results). Requires sast/architecture.md (run sast-analysis first). Outputs findings to sast/fileupload-results.md. Use when asked to find file upload, unrestricted upload, or extension bypass bugs.
Provisions a Replay Vision `experiment` scanner that watches one experiment’s exposed sessions, records each session’s variant, samples variants evenly, and stops when the experiment ends. Covers status guards, a prompt that remains comparable across variants, sizing credit spend against the experiment’s own population, creating it disabled so its prompt can be previewed on real sessions, and reading results by variant. TRIGGER when: the user wants Replay Vision to watch an experiment, asks to scan or analyze an experiment’s recordings with AI, asks “what are users actually doing in the test variant,” or wants a scanner scoped to an experiment’s exposed sessions. DO NOT TRIGGER when: creating a general-purpose scanner not tied to an experiment (use creating-replay-vision-scanners), reading observations a scanner already produced (use exploring-replay-vision-observations), or manually browsing an experiment’s recordings without AI analysis (use analyzing-experiment-session-replays).
Standard single-cell RNA-seq analysis pipeline. Use for QC, normalization, dimensionality reduction (PCA/UMAP/t-SNE), clustering, differential expression, and visualization. Best for exploratory scRNA-seq analysis with established workflows. For deep learning models use scvi-tools; for data format questions use anndata.
Recompute and display the two database scores (Design & Integrity + Performance & Scale) from a findings JSON, without re-auditing. Use to re-show or refresh the scores after an audit, to re-score with a different paradigm profile, or to score a saved findings file.
Scrape pages locally with anti-bot bypass, TLS impersonation, and adaptive element tracking — no API keys, no cloud. Handles Cloudflare protection, CSS/XPath element extraction, and survives site redesigns. Complements firecrawl (cloud) with 100% local execution. Triggers on Cloudflare bypass, anti-bot scraping, stealth fetch, local scraping, Scrapling.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.