agent-browser wraps playwright for ai-driven web automation, exposing six functions (open, snapshot, interact, screenshot, execute, close) with element refs and session persistence for multi-step workflows.

designing-surveys provides tactical guidance on survey construction grounded in product research methodology, emphasizing csat over nps and single-variable question design for measuring satisfaction and feature prioritization.

establishes 10x-cli on a machine by checking install state, fetching the current readme, and following its prescribed setup path. delegates version and dependency specifics to the source-of-truth readme rather than hardcoding.

sveltekit-data-flow provides file placement guidance and basic patterns for load functions, form actions, and server vs client data flow, with limited depth on serialization and invalidation.

roier-seo runs lighthouse audits on websites or localhost and surfaces performance, accessibility, seo, and core web vitals scores. includes templated fixes for common issues like meta tags and structured data.

openduck-distributed-duckdb provides distributed query execution across local and remote duckdb workers via transparent attach syntax, with append-only differential storage and hybrid plan splitting via grpc.

pulumi-best-practices covers dependency tracking, component design, secret management, and refactoring patterns for infrastructure code. diagnostic signals included but procedures remain incomplete.

apify-generate-output-schema analyzes actor source code to produce dataset, key-value, and output schemas, with emphasis on data validation and cross-checking against runtime code.

testing-dags automates iterative test-debug-fix cycles for airflow dags with diagnostic output and validation commands, supporting timeouts and retries across multiple failure scenarios.

golang-project-layout guides developers through sizing go project structure to actual needs, starting with architectural questions before applying patterns like dependency injection and 12-factor principles.

trigger-tasks provides durable background job execution with retry logic, queuing, scheduling, and multi-tenant isolation. covers task creation, schema validation, concurrency management, cron scheduling, and real-time observability via trigger.dev sdk.

mapkit-location covers ios 17+ map and location features using swiftui mapkit, async corelocation apis, geofencing, geocoding, and local search with a review checklist for common gotchas.

flutter-apply-architecture-best-practices covers layered application structure via mvvm, repository injection, and separation of concerns. incomplete treatment of logic and data layers limits actionability.

videoagent-video-studio generates short ai videos from text or images across seven backend models without requiring api keys. it automatically selects generation mode, enhances prompts, and supports configurable duration and aspect ratios.

constitutional ai trains models to critique and revise their own responses using principle-based guidance, eliminating the need for human labels in safety alignment through self-supervised learning and ai preference ranking.

indicator-chart generates interactive plotly charts for 100+ technical indicators across multiple exchanges. supports overlay and subplot layouts with signal markers and plain-language interpretations of current readings.

clickstack-otel-collector guides users through two distinct paths to integrate opentelemetry collectors with managed clickstack on clickhouse cloud, either by deploying a new local collector or configuring an existing one with the clickhouse exporter.

dbs-restore retrieves the most recent diagnostic snapshot for a project, presenting saved state (conclusions, rejected directions, pending assumptions) so users can resume work. integrates with dbs-save for persistent session storage across diagnostic workflows.

llm-evaluation catalogs metric types (automated, human, llm-as-judge) and provides a python skeleton for running evaluation suites, but lacks deployment guidance, failure recovery, and concrete validation workflows.

browser-extension-builder covers manifest v3 architecture, content scripts, background workers, and storage patterns. includes project structure templates and anti-patterns but lacks publishing workflows and cross-browser specifics.

golang-concurrency provides guidance on goroutine patterns, leak prevention, and structured concurrency auditing across write, review, and codebase audit modes. emphasizes ownership and context propagation.

asc-cli-usage documents discovery patterns and flag inspection for app store connect command design, covering help systems, search, schema inspection, and capability checking.

cmux-shared-behavior establishes mutation-path rules for behaviors exposed across multiple entrypoints, enforcing single code paths, optimistic update reconciliation, and test coverage alignment with real bug reports.

operates adobe desktop apps (photoshop, illustrator, premiere, after effects, indesign, audition) via flue bridge, piping extendscript through stdin/stdout for bounded document inspection and editing.

npm-package covers bun-based library authoring, build setup, and npm publishing. addresses esm/cjs interop and exports configuration for single-package repos.

kafka-development covers architectural principles, topic design, producer/consumer best practices, schema management, and distributed patterns like event sourcing and saga coordination. practical but lacks executable procedures.

manim-video generates technical explainer animations for graphs, workflows, and system diagrams using motion and progressive reveal. outputs 16:9 mp4 with storyboard and thumbnail, then hands off to wider video stack if needed.

query-token-info exposes four read-only token endpoints (search, meta, dynamic, kline) via a unified cli, supporting ethereum, bsc, base, and solana. returns metadata, market data, and candlestick ohlcv across configurable time windows.

co-marketing covers partner identification, campaign brainstorming, and execution frameworks for saas companies seeking joint marketing opportunities. includes scoring criteria, outreach templates, and post-campaign measurement.

pixijs-events covers the federated event system for pointer, mouse, touch, and wheel input in pixi.js v8, with eventmode configuration and global/local coordinate tracking for interactive scene graph objects.

reasoningbank-with-agentdb enables adaptive learning for agents via vector-backed pattern storage, trajectory tracking, verdict judgment, and memory distillation with claimed 150x-500x performance gains over legacy implementations.

archify generates interactive, self-contained html diagrams from json specifications across architecture, workflow, sequence, dataflow, and lifecycle domains. static output is default; motion appears on request.

zod covers 43 validation rules across schema definition, parsing, type inference, and error handling. organizes patterns by priority level and references external rule files for detailed implementation guidance.

scoping-cutting aggregates frameworks from product leaders on appetite-driven scope management, feature cutting, and hypothesis validation. provides structured questions and common pitfalls but lacks concrete decision logic for agents.

styled-components-best-practices documents css-in-js patterns for react, covering theming, responsive design, animations, and performance optimization through code examples and structural guidance.

google-calendar provides command-line access to google workspace calendars with oauth authentication and basic crud operations, event discovery, and attendee management. no mcp server required.

copilotkit-contribute guides developers through forking, building, and submitting pull requests to the copilotkit monorepo using pnpm, nx, and conventional commits. covers workflow from local setup through ci validation.

testing covers sql and rust test formats, execution methods via make and cargo commands, and patterns for writing .sqltest files with declarative syntax and expected result assertions.

stock-analysis aggregates sentiment data and blogger viewpoints for individual equities, compiling bull/bear arguments and position data into a structured report without investment recommendations.

role-based access control mixin for caffeine ai apps. provides admin/user/guest roles, internet identity integration, and attribute capture (name, email). first login auto-promotes to admin.

hyperframes-cli scaffolds video compositions with init templates, lints for structural errors, previews with hot-reload, and renders to multiple formats. environment checks and audio tasks delegate to hyperframes-media.

scrape fetches clean markdown, html, json, or screenshots from single or batched urls via bright data cli, with block detection and escalation to browser automation for javascript-heavy content.

design covers brand identity, logos, corporate identity programs, presentations, banners, social media assets, and icons via python scripts and external sub-skills. routes tasks to specialized handlers with ai generation and html export workflows.

extension-data-viewer provides auto-generated admin queries over stable canister state. supported types (map, set, list variants) expose paginated read-only endpoints that trap on non-controller access. requires no additional setup beyond template defaults.

propose-hypotheses executes a multi-stage first principles framework workflow: hypothesis generation, logic verification, evidence validation, trust audit, and final decision. relies on external fpf-agent with file-based state passing between numbered steps.

codebase-search provides a structured workflow for navigating code via semantic search, grep patterns, and glob file discovery. covers function tracing, dependency mapping, and impact analysis with concrete language-specific examples.

publishing-press-publish handles CLI publication to the printing-press-library repo via pull request, with strict guardrails against publishing incomplete work and a separate blocked-api journal mode for recording known blockers.