Skills
98,067 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Builds evidence-traceable market research reports and assumption-driven market sizing or forecast scenarios. Use for market definition, industry and customer evidence, competitive landscapes, TAM/SAM/SOM reconciliation, forecast sensitivity, and auditable report scaffolds.
Find and verify external literature, prior art, datasets, benchmarks, and citation candidates. Use for 文献检索, 相关工作, benchmark搜索, and research opportunity maps. Recurring recent-paper watch belongs to ccf-literature-monitor; existing-citation audits, manuscript assessment, and result schemas have separate owners.
Create an implementation task plan in `_/local-plans/<plan-name>.md`. Use when asked to make a plan, design, consider an implementation approach, or break down tasks. First investigate the codebase with Explore Agent, then describe work in verifiable, parallelizable units. Follow the verifiable standard format from plan-verifier Agent (background, current state, design, file structure tree, implementation steps, verification method). For implementation, see implement-issue; to create Issues, see create-issue-tree (or create-issue if it is not installed).
Diagnoses and resolves Filestore client mount failures on Google Cloud, permission errors (EACCES), and network timeouts (ETIMEDOUT). Use when an NFS mount hangs or fails from a Compute Engine VM, GKE pod, Cloud Run service, or Vertex AI workload, when `mount.nfs` reports "Connection timed out" or "access denied by server", when checking whether VPC ingress firewall rules or `nfsExportOptions` allow a client IP, or when a previously working Filestore share suddenly stops mounting after an administrative change. Don't use for Cloud Storage (GCS) buckets, Persistent Disk, or Cloud NetApp Volumes, and don't use for Filestore capacity scaling or backup and export-policy auditing.
Diagnose and reverse traffic loss on existing pages. Use when the user asks about content decay, pages losing traffic, declining rankings, traffic drops, why a page stopped ranking, content refresh strategy, or when to consolidate vs redirect old content. For creating new content, see brief.
Upgrade a vigiles spec's guidance() rules to enforce() — scan the guidance rules in a CLAUDE.md/AGENTS.md spec and find existing linter rules (ESLint, Ruff, Clippy, Pylint, RuboCop, Stylelint) that back them. Use when asked to strengthen, harden, or make vigiles rules enforceable; NOT for general linting or fixing lint errors.
GCP cloud resources including Compute Engine, GKE, Cloud Run, Pub/Sub, VPC networking, DNS, IAM, Secret Manager, and monitoring. Monitor GCP infrastructure, analyze resource usage, audit security posture, and manage organizational hierarchy across projects and folders.
Write production-grade Playwright tests in TypeScript: Page Object Model, fixtures, auto-waiting, user-facing locators, parallel execution, CI integration, sharding, and installed-version checks. Includes an explicit "do not" list for AI agents. Use when: "Playwright," "write E2E test," "page object," "new Playwright suite," "Playwright config." Not for: fixing one flaky test at runtime — use test-reliability. Not for: bulk regenerating selectors after a UI refactor — use selector-drift-recovery. Not for: visual baseline creation/management — use visual-testing. Not for: deep WCAG/axe audits — use accessibility-testing. Related: visual-testing, ci-cd-integration, api-testing, test-reliability, selector-drift-recovery, accessibility-testing.
Design and interpret CCF experiments: datasets, baselines, metrics, ablations, execution specifications, observed feedback, and result-table contents. Use for 设计实验, 消融, benchmark planning, 实验结果分析, and evidence design within requested autonomous research. Preserve real values. Orchestrator coordinates research-loop execution; visual styling and broad retrieval have separate owners
Beisen job management. Used to search recruiting jobs, query job lists by job title/code/status/recruiting category, and view job details, JD, responsibilities, qualifications, headcount, salary range, work location, recruiting process and application counts. Use when the user asks about jobs, recruiting positions, JDs, job responsibilities, job requirements or job status, or needs to locate a jobId first for other recruiting skills.
Use when looking up library documentation, API references, framework patterns, or code examples for any library (React, Next.js, Vue, Django, Laravel, etc.) and the Context7 MCP server is unavailable or not configured. Fetches the same current docs directly via the Context7 REST API as a fallback. Triggers on: how to use library, API docs, framework pattern, import usage, library example
Use when writing blog posts or documentation markdown files - provides writing style guide (active voice, present tense), content structure patterns, and MDC component usage. Overrides brevity rules for proper grammar. Use nuxt-content for MDC syntax, nuxt-ui for component props.
Beisen recruiting to-dos. Used to summarize my recruiting items, view the number of resumes pending screening, interviews pending evaluation and interviews pending attendance, confirm upcoming interview schedules, and display the current user's recruiting to-do list. Use when the user asks what recruiting to-dos they have, about resumes pending processing, interviews pending evaluation, whether there are interviews today or soon, or about interview schedules.
Guide AI agents through Godot 4.x GDScript coding best practices including scene organization, signals, resources, state machines, and performance optimization. This skill should be used when generating GDScript code, creating Godot scenes, designing game architecture, implementing state machines, object pooling, save/load systems, or when the user asks about Godot patterns, node structure, or GDScript standards. Keywords: godot, gdscript, game development, signals, resources, scenes, nodes, state machine, object pooling, save system, autoload, export, type hints.
Detects host inventory and effective CPU, memory, disk, scheduler, container, and accelerator limits when a user asks for resource-aware planning or before a clearly resource-sensitive local workload. Produces a redacted JSON snapshot and conservative planning helpers without stress tests or assuming visible host hardware is usable.
Beisen recruiting insights. Used to view overall recruiting progress, locate process bottlenecks, view progress on key jobs, review stage completion volumes, track Offer-to-onboarding conversion, and launch interview quality analysis or competitor intelligence analysis, and view or cancel analysis asynchronous tasks. Use when the user asks about recruiting progress, interview progress, stage completion volumes, process bottlenecks, key job progress, Offer-to-onboarding conversion, interviewer question quality, interviewer comparisons for the same job, examples of excellent questions, inefficient interview questions, competitor intelligence, or stopping an analysis task.
Runs Bright Data jobs from a terminal with the bdata CLI, the published npm package @brightdata/cli. Use when the agent itself needs to run a Bright Data job from a shell right now, when building or testing a scraper, when a quick one-off check of a page, a search, a zone or a bill is enough, or when the user names bdata, brightdata, or the Bright Data CLI. Not for deterministic code that repeats the same job on a schedule, which belongs to brightdata-sdk. Not for an AI app that picks its tools at run time, which belongs to brightdata-mcp. Not for a first-time install or a login that will not go through, which belongs to agent-onboarding.
Explore and understand PUDL energy data: discover which tables exist, look up column meanings and usage warnings, and load Parquet files from S3 or a local directory. No PUDL Python package required. Use this skill whenever a user asks what PUDL data contains, wants to understand a specific table or column, asks about data quality or limitations, needs help loading data into a notebook or script, or wants to know which table covers a topic like electricity generation, utility financials, fuel costs, power plant locations, emissions, capacity factors, FERC financial data, or EIA survey data. Also use when the user mentions PUDL, Catalyst Cooperative energy data, or any of the specific data sources PUDL ingests (EIA-860, EIA-861, EIA-923, FERC Form 1, FERC Form 714, FERC EQR, EPA CEMS, EPA CAMD, etc.).
Beisen recruiting demand management. Used to query recruiting demand lists and details, filter recruiting demands by demand status/name/number/requester, and view HC/headcount, demand owner, expected onboarding time, associated jobs, and demand-level progress such as Offer/pending onboarding/onboarded. Use when the user asks about recruiting demands, HC, demand details, demand status, demand owner, how many people a demand still needs, or demand onboarding progress.
Beisen talent sourcing. Used to search talent pool resumes by job requirements, query AI-recommended talent for a job, batch-fetch recommended candidates by job, poll or cancel asynchronous talent sourcing tasks, and follow up on candidate profile analysis leads. Use when the user asks to find resumes, search the talent pool, recommend candidates, source by job requirements, recommend resumes for a job, or check talent sourcing task progress.
Assess research ideas for value, novelty, insight, and mechanism. Use for 思路审核, 靠谱吗, 值得做吗, 创新够不够, idea review, scoring, and ranking; no numeric-score request is needed. Default to concept-only review without experiment assessment, including ideas extracted from manuscripts. Developing an idea belongs to ccf-idea-optimizer; evaluating manuscript evidence or writing belongs to ccf-paper-reviewer.
Swift iOS architecture guidance and playbooks for MVVM, MVI, TCA, Clean Architecture, VIPER, MVP, Coordinator, and Reactive patterns. Use when designing, implementing, refactoring, or reviewing the architecture of a SwiftUI or UIKit feature, module, or codebase.
Scan recent arXiv/OpenReview papers, venue feeds, labs, and competitors for overlap. Use for 竞品监控, 新论文追踪, 论文跟踪, and recurring novelty watch. Give dated evidence and changes since the prior scan. Broad related-work retrieval belongs to ccf-literature-searcher.
Install vigiles and test a Claude Code harness — hooks, skills, agents, settings, CLAUDE.md — by picking the right tier (unit / deterministic / eval) and writing a test that passes. Use to check that a hook fires or blocks, that a skill triggers, that injected context lands; or to observe a run — which tools it called, whether it stayed inside its declared allowed-tools, what files and side effects it produced, how to intercept a call without executing it.
Remove signs of AI-generated writing from text. Use when the user mentions 'sounds like AI,' 'too robotic,' 'humanize this,' 'make it sound human,' 'de-AI this,' 'sounds like ChatGPT,' 'make it natural,' or 'doesn't sound like a person wrote it.' Detects and fixes AI writing patterns including inflated significance, promotional language, formulaic structure, AI vocabulary, superficial analyses, vague attributions, hedging, robotic rhythm, em dash and boldface overuse, chatbot artifacts, and missing personality. Based on Wikipedia's "Signs of AI writing" guide and editorial best practices.
Read a Prisma Next structured error envelope and route to the right recovery — code, domain, severity, why, fix, meta. Use for error, exception, my emit failed, my query won't typecheck, my query crashed, my migration won't apply, MIGRATION.HASH_MISMATCH, BUDGET.ROWS_EXCEEDED, BUDGET.TIME_EXCEEDED, RUNTIME.ABORTED, PLAN.HASH_MISMATCH, CONTRACT.MARKER_MISSING, PN-RUN-3001, PN-RUN-3002, PN-RUN-3030, PN-MIG-2001, PN-CLI-4011, PN-SCHEMA-0001, drift, capability missing, planner conflict, prisma studio, EXPLAIN, query log, db.end, db.close, script won't exit, hangs, close connection, pool.end, client is closed.
Review what Prisma Next migrations will run on merge or deploy, render the migration graph, resolve concurrent / diamond-convergence conflicts, and configure environment refs for CI. Use for "what migrations are going to run", "what runs on deploy", merge conflict, diamond convergence, concurrent migrations, migration status, ref management, staging, production, MIGRATION.DIVERGED, MIGRATION.NO_MARKER, MIGRATION.MARKER_NOT_IN_HISTORY, prisma migrate status, prisma migrate diff, prisma migrate resolve.
Wire the Prisma Next runtime — `db.ts` setup using `postgres<Contract>(...)` from `@prisma-next/postgres/runtime`, `sqlite<Contract>(...)` from `@prisma-next/sqlite/runtime`, or `mongo<Contract>(...)` from `@prisma-next/mongo/runtime`; middleware composition (telemetry from `@prisma-next/middleware-telemetry`; lints and budgets), `DATABASE_URL` config, per-environment branching, switching between Postgres, SQLite, and Mongo façades. Use for db.ts, postgres(), sqlite(), mongo(), middleware, telemetry, lints, budgets, DATABASE_URL, .env, connection pool, poolOptions, dev vs prod config, transactions, db.transaction, read replicas, multi-database, script won't exit, hangs, close connection, db.end, db.close, pool.end, [Symbol.asyncDispose], await using.
Explore and query any dataset annotated with a Frictionless Data Package descriptor (datapackage.json). Use this skill whenever a user wants to discover what tables or resources a dataset contains, look up column names and descriptions, surface usage warnings embedded in metadata, or understand how to load data from Parquet files, DuckDB or SQLite databases, or CSV files described by a datapackage.json. Also use when the user has a datapackage.json and wants to know what's in it, how to query it efficiently, or how to connect its metadata to actual data files. Pairs well with dataset-specific skills (like `pudl`) that layer domain knowledge on top
Build reusable, composable n8n sub-workflows. Use when extracting shared logic, building anything multi-step or reused across workflows, or any workflow over ~10 nodes — and whenever the user mentions sub-workflows, Execute Workflow, reuse, shared/common logic, modular workflows, "Define Below" inputs, waitForSubWorkflow, mode each vs all, or exposing a workflow as an agent tool. Covers typed sub-workflow inputs, all-vs-each execution, verb-first naming for discovery, stateless vs stateful design, and splitting by input shape.
Apple Human Interface Guidelines for iPhone. Use when building, reviewing, or refactoring SwiftUI/UIKit interfaces for iOS. Triggers on tasks involving iPhone UI, iOS components, accessibility, Dynamic Type, Dark Mode, or HIG compliance.
Deploy a production self-hosted n8n end to end to a fresh Linux VM over SSH, using Docker Compose behind a Caddy reverse proxy with automatic HTTPS. Use whenever the user wants to self-host, install, provision, or deploy n8n on their own server/VPS (Hetzner, DigitalOcean, AWS EC2, bare metal) — in single/regular mode or queue mode with workers — or to update, back up, restore, or harden such an instance, or make Python Code nodes run on it (task runners). For SELF-HOSTED n8n (Docker), not n8n Cloud and not building workflows. The skill makes the agent ask single-vs-queue first, collect domain/SSH/timezone inputs, generate fresh secrets on the box, and bring the stack up with TLS. Trigger on “deploy n8n”, “self-host n8n”, “n8n docker compose”, “n8n queue mode / workers”, “n8n reverse proxy / SSL”, “back up / update my n8n”, “Python runner unavailable” / “n8nio/runners sidecar”, or “we don't want to give every user the OAuth client secret” / “enable Sign in with Google” (credential overwrites).
Manage durable working-session memory for coding agents. Use when a user asks to preserve or recover agent context across disconnects, VS Code restarts, long-running work, handoffs, or any session where important state should be written periodically under the repo's session directory. Do NOT use for: simple questions, short tasks, one-off commands, linting, or code review.
GPU-accelerates scientific Python on NVIDIA hardware and verifies that the result is correct and faster. Use for CUDA/GPU optimization; CPU-bound NumPy, SciPy, pandas, scikit-learn, NetworkX, scikit-image, vector-search, image-processing, graph, simulation, or file-I/O workloads; CuPy, cuDF, cuML, cuGraph, cuVS, cuCIM, KvikIO, Warp, Newton, Numba-CUDA, or RAFT questions; and profiling, memory-transfer, kernel, or multi-GPU bottlenecks. Also use when large data-parallel Python code is slow and GPU acceleration is a plausible option, even if the user does not name CUDA.
Wire Prisma Next into the project's build system with the right build-tool plugin — Vite today via @prisma-next/vite-plugin-contract-emit (Vite 7 / 8); Next.js / Webpack / esbuild / Rollup / Turbopack are named as gaps rather than fabricated. Always offers the Vite plugin proactively when the project is using Vite. Use for vite plugin, vite-plugin, vite.config.ts, prismaVitePlugin, contract emit on save, HMR, hot reload contract, dev server, Next.js plugin, next plugin, withPrismaNext, webpack plugin, esbuild plugin, rollup plugin, build integration, dev server plugin, vite 7, vite 8.
Drafts overdue-invoice reminder emails from the ledger (MYOB, NetSuite, QuickBooks, Xero, or Zoho Books) plus PayPal, Stripe, and Airwallex data, matched to each customer's payment history and tone (gentle for good customers, firm for repeat late payers). Sends via PayPal with owner approval; everything else queues as a mail draft, with the Airwallex hosted pay link included where one exists. Use when the user asks \"who owes me money,\" mentions overdue invoices, or wants to follow up on unpaid invoices.
Use when handling files, images, attachments, or binary data in n8n, OR when an AI agent needs to take a user-uploaded file as tool input or return a generated file. For Data Tables (schemas, dedup, persistent state), see the separate n8n-data-tables-official skill. Triggers on "file", "image", "PDF", "attachment", "binary", "upload", "download", chat trigger with files, agent tool that needs a file, vision/multimodal, or any handling of non-JSON file data.
Builds and checks Bayesian models with PyMC, including hierarchical models, NUTS MCMC, variational inference, mutable-data predictions, posterior predictive checks, diagnostics, and PSIS-LOO model comparison. Use for probabilistic modeling and uncertainty inference in PyMC.
Render and redesign scientific figures, visual tables, and method/architecture diagrams from supplied content, including biological data. Use for 绘图美化, 排版, 配色, GPT Image 2 generation, pure SVG, and editable SVG/PDF/PPTX. Preserve values and topology. Experiment evidence design belongs to ccf-experiment-designer; manuscript rewriting and PDF-to-writing exemplars are separate
Edit the Prisma Next data contract — add models, fields, relations, indexes, enums, value objects (composite types), type aliases, namespaces (Postgres schemas), cross-contract foreign keys (cross-space FK), polymorphic types (`@@discriminator` / `@@base`), use extension namespaces (`pgvector.Vector(...)`, `cipherstash.EncryptedString(...)`), wire `prisma-next.config.ts` with `defineConfig` from the `@prisma-next/<target>/config` façade, and run `prisma-next contract emit`. Use for schema, models, fields, attributes, soft delete, paranoid, scopes, validations, callbacks, prisma schema, PSL, contract.prisma, contract.ts, contract.json, contract.d.ts, `@prisma-next/postgres/config`, `@prisma-next/postgres/contract-builder`, `@prisma-next/postgres/control`, `@prisma-next/mongo/config`, `@prisma-next/mongo/contract-builder`, `extensions:`, pgvector, cipherstash, postgis, paradedb, supabase, `@prisma-next/extension-supabase`, `@@control`, control policy, managed, tolerated, external, observed.
Convert web articles or blog posts into slides through a content.json pipeline, translate or rebuild existing PPTX files onto a template, and edit an already-open PowerPoint deck with COM Automation (RefURL, overflow review). Use for article-to-slides conversion, deck translation, template-based rebuilds, or live COM editing of an open deck. Not for simply reading, inspecting, or lightly editing a .pptx file. Triggers on 記事をスライド化, スライド翻訳, 開いているパワポを編集, COM自動化, RefURL, content.json.
Audit existing CCF claims, numbers, terminology, and citations against supplied or verified evidence. Use for 引用核验, claim审计, 数字一致性, and BibTeX/context checks. Full scientific review belongs to ccf-paper-reviewer; new literature discovery belongs to ccf-literature-searcher.
Review dependency PRs with structured research, existing-PR-discussion capture, multi-lens analysis (security, code quality, impact), and a repeatable verdict template. USE FOR: dependency update PRs, Renovate/Dependabot PRs, library upgrade reviews, "review this dependency PR", "should we merge this update". DO NOT USE FOR: feature PRs, application code reviews, dependency automation/bot configuration, or unattended merge without confirmation
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.