knowledge-agent builds filtered observation corpora and loads them into conversational ai sessions for domain-specific querying. users filter observations by project, type, concept, or date, then prime the corpus and ask questions conversationally.

coingecko exposes spot prices, ohlc candles, market discovery, and global crypto stats via a curated python function library. includes extensive decision trees and tool-mapping guidance but lacks procedural depth for real-world integration.

web-artifacts-builder provides a scripted workflow for bundling react typescript projects with tailwind and shadcn/ui into self-contained html artifacts, with opinionated tooling and design guardrails to avoid common ai-generated output patterns.

requesting-code-review dispatches a focused code reviewer subagent with git context (base/head SHAs) to catch issues before merge, emphasizing early and frequent review across task-driven development workflows.

eval ranks agent results across a session using configurable metrics, git diffs, or llm judgment. supports metric mode, judge mode, and hybrid evaluation with tie-breaking logic.

context-compression provides three production-ready compression strategies for managing multi-million token agent sessions. it contrasts naive token minimization against task-token optimization, establishing compression-quality evaluation frameworks and offering anchored summarization and opaque approaches with explicit tradeoffs.

solana-dev covers end-to-end dapp development including react/next.js ui layers, anchor program writing, wallet integration, and local testing strategies. integrates solana mcp for live documentation and error diagnosis.

create-skill teaches skill authorship conventions for the implexa framework, emphasizing progressive disclosure, structural constraints (200-line rule), and modular packaging patterns for agent capability extensions.

service-mesh-observability covers metrics, traces, and logs across istio and linkerd with concrete prometheus queries, jaeger setup, and dashboard templates. practical but procedurally thin on implementation sequencing and failure recovery.

nextjs-app-router-patterns covers server/client rendering modes, file conventions, and architectural decisions for next.js 14+ applications. includes layout setup, data fetching with revalidation, and best practices for server components and parallel routes.

gh-review-requests queries github for unread review notifications on open prs filtered by team membership or authorship, displaying results in a markdown table with fallback manual api queries.

design-flow orchestrates a seven-phase guided workflow from requirements gathering through frontend build and optional review. it sequences dependent skills and manages state across phases, but lacks explicit failure handling and assumes all upstream skills exist and function correctly.

diagnose-seo provides a four-layer framework for identifying crawlability, indexability, renderability, and signal issues. organized by dependency chain with specific checks per layer, but lacks concrete input requirements and automated detection logic.

pr creates a pull request to the canary branch after gathering git context, handling uncommitted changes, pushing, searching related issues, and populating a pr template with gitmoji-formatted titles and issue links.

vercel-deploy packages projects into tarballs, auto-detects frameworks, and uploads to vercel for instant deployment, returning preview and claim urls for site access and account transfer.

output-sanitizer redacts leaked credentials, pii, and internal paths from agent responses using regex patterns and classification rules. detects aws keys, api tokens, emails, phone numbers, database urls, and source code secrets with reporting.

postman-collection-generator automates collection creation from express, next.js, fastify, hono, and nestjs route definitions via ast parsing and regex scanning, outputting v2.1 json with variables and folder organization.

article-writing guides long-form content creation with emphasis on concrete examples, tight prose, and authentic voice. it covers blog posts, essays, guides, and newsletters with rules against filler and fake credibility.

sanity covers schema design, groq query patterns, typescript integration, and project layout for headless cms work. lacks trigger phrases, decision trees, and failure recovery guidance.

grafana-dashboards covers dashboard design principles, panel configuration patterns, and provisioning workflows. provides json templates for api, infrastructure, and database monitoring but lacks numbered procedures and failure recovery guidance.

ui-visual-validator codifies a skeptical visual testing methodology emphasizing evidence-based assessment, accessibility compliance, and systematic cross-platform validation. role-based expert for design system and regression testing verification.

momentic-result-classification uses Momentic MCP tools to analyze failing test runs, categorize failures into predefined buckets, and determine recoverability (recoverable, one-time recoverable, non-recoverable).

troubleshooting-dbt-job-errors guides systematic diagnosis of dbt cloud job failures through log analysis, api querying, and data investigation. covers when to use and the principle of investigating root causes before modifying tests.

brainstorm-experiments-new guides hypothesis formation and lean experiment design for product validation. it structures thinking around xyz hypotheses and suggests low-effort pretotype methods (landing pages, videos, waitlists, manual mvps) grounded in skin-in-the-game principles.

skill-installer manages installation of codex skills from curated openai lists or custom github repositories, with support for public and private repos via git or direct download.

hive-create-task guides users through a six-phase wizard to design, scaffold, test, and publish benchmark tasks for agent evaluation. emphasizes eval design quality and produces a complete, runnable task repo.

crypto-agent-trading enables purchase, sale, and swap operations against crypto.com via typescript api clients, with quote-confirm flows, balance queries, coin discovery across 200+ assets, and optional auto-execution via persistent memory.

senior-mobile covers scaffolding and optimization for react native, flutter, and native ios/android development. includes platform decision guidance, workflow examples for swiftui and jetpack compose, and app store submission tooling. static analysis focus; omits runtime profiling and ci/cd.

project-skill-audit reviews past codex sessions and local skill folders to recommend which skills to create or refine, grounded in actual project workflows rather than speculation.

svelte-runes covers reactive state, computed values, side effects, and component props using svelte 5 rune syntax. references guide migration from svelte 4, common mistakes, and when to apply each rune.

android-data-layer teaches repository pattern basics with room and retrofit, establishing local-first data flow via coroutines. covers architecture, not production sync complexity.

getting-started-with-skills documents mandatory workflows for skill discovery and usage within the implexa environment, emphasizing read-before-use discipline, todowrite integration for checklists, and explicit announcement patterns to maintain process transparency.

play-developer-console wraps a local bun typescript cli for managing google play store metadata, images, release tracks, and in-app products. targets moshi android releases with keyless gcloud auth.

auth0-migration provides a phased workflow for switching existing authentication systems to Auth0, covering user export, bulk import, code updates, and gradual cutover strategies.

netlify-deploy automates site deployments via the netlify cli, handling authentication, framework detection, and site linking. supports oauth and api tokens with error recovery for common failure modes.

react-ui-patterns establishes core principles for handling loading states, errors, and async data in react components. covers decision trees for common UI patterns and documents anti-patterns to avoid.

rust-async-patterns covers production-ready concurrent patterns for tokio-based applications, including task coordination, channels, error handling, and debugging with tokio-console. teaches when to apply patterns like joinset, broadcast channels, and cancellation tokens in real systems.

code-connect-components bridges figma design components to code implementations via code connect, automating component discovery and mapping with manual review gates across multiple frameworks and languages.

network-bgp-diagnostics provides read-only troubleshooting patterns for neighbor state transitions, prefix exchange validation, and policy filtering inspection without operational changes.

multiplayer-game documents netcode patterns, physics selection, and security baselines across 10 game archetypes. covers matchmaking, tick loops, state replication, and interest management with starter code, though lacks step-by-step procedures and specific trigger phrases.

sponsor-finder identifies sponsorable dependencies in a github repository by querying the deps.dev api for dependency graphs and funding metadata, then validates and groups results by sponsorship destination.

captures architectural decisions during development as structured ADR documents. auto-detects decision moments from conversation patterns, formats them using Nygard's template, and enables querying past rationales.

swiftui-performance-audit diagnoses rendering slowness, janky scrolling, and memory issues via code review and instruments profiling. it classifies symptoms, applies anti-pattern checks, and guides trace analysis with remediation patterns.

postgres-patterns provides reference material on index selection, data types, and sql patterns for optimization and security, but lacks procedural depth and explicit failure handling guidance.

planner generates phased implementation roadmaps starting with requirement clarification and codebase research, decomposing work into independent sprints with atomic tasks and documented validation methods.

ton-vulnerability-scanner identifies three security patterns in ton smart contracts (integer-as-boolean misuse, fake jetton handlers, unsafe gas forwarding). detects funC code via file extensions and project structure, then surfaces vulnerable code with pocs and fix recommendations.

mkdocs-translations automates bulk documentation translation into target languages while preserving markdown structure, folder hierarchy, and mkdocs configuration. requires explicit language/locale specification upfront and mirrors en directories into locale-specific branches.