caveman
claudeTalk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
AI engineering methodology spine + 160+ skill library, activated at session start. Install once; the agent uses it natively.
Plugin aix from marketplace aix in truongnat/aix. Commands read from the repo's manifest — 7 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
3d-motion-pro\"3D design and motion graphics for the web: Three.js scenes, WebGL shaders, GLTF model loading, particle systems, physics simulations, AR/VR with WebXR, and GSAP animation pipelines.\"\nUse this skill when building interactive 3D web experiences, loading and optimizing 3D models for browsers, imple
a11y-design-pro\"Accessible and inclusive design: WCAG 2.1/2.2 compliance, ARIA patterns, keyboard navigation, screen reader optimization, color contrast, semantic HTML, and automated accessibility testing pipelines.\"\nUse this skill when auditing or implementing accessibility for web UIs, adding ARIA to dynamic
a2a-protocol-pro\"Design and implement multi-agent systems using the Agent-to-Agent (A2A) protocol: agent discovery, task delegation, lifecycle management, and security.\"\nUse this skill when building systems with multiple collaborating agents, designing agent discovery (Agent Cards), implementing task delegation
accounting-proUse this skill whenever the user wants to perform accounting tasks, bookkeeping, financial statement preparation, or work with accounting systems. This includes double-entry bookkeeping, financial statement preparation (balance sheet, income statement, cash flow statement), journal entries, account
ag-ui-pro\"Build interactive user interfaces for AI agents using the AG-UI protocol: streaming chat, human-in-the-loop, generative UI, shared state, and frontend tool calling.\"\nUse this skill when designing agent-interactive frontends, implementing streaming chat UIs, adding human approval gates, building
agent-desktopDesktop application automation for AI agents using vercel-labs/agent-desktop. Reads native OS accessibility trees (macOS AX API, Windows UI Automation, Linux AT-SPI) and executes UI actions via stable snapshot-scoped refs. Use when an agent needs to see, navigate, and operate desktop software withou
agent-evaluation-proEvaluate, observe, and improve AI agent / LLM output quality with structured metrics, regression testing, red-teaming, and CI/CD integration.\nUse this skill when measuring agent correctness, setting up prompt regression tests, debugging agent failures at scale, or building an evaluation pipeline fo
ai-agents-proExpert design and orchestration of autonomous AI agents, multi-agent systems, tool use, and persistent memory.
ai-design-pro\"AI-enhanced design practices using generative tools, AI assistants, and automated workflows: image generation, rapid prototyping, AI-driven UX decisions, motion graphics, photorealistic visuals, and automated image processing.\"\nUse this skill when working with AI design tools (Midjourney, DALL-E
ai-integration-pro\"Professional AI/LLM integration: Claude API, OpenAI API, prompt engineering, streaming, tool use/function calling, multi-turn conversations, RAG, embeddings, conversation state, evaluation, routing, observability, and production reliability patterns.\"\nUse this skill when the user integrates LLMs
ai-red-teaming-proAdvanced security testing for AI systems, specializing in prompt injection, jailbreaking, and model robustness.
algorithm-pro\"Professional algorithm engineering: formal modeling, problem taxonomy (optimization, counting, online, …), strategy selection (greedy/DP/graph/search), correctness and complexity proofs, pattern recognition, competitive debugging (WA/TLE/MLE), systems constraints (memory, IO, streaming), and perfo
android-proExpert Android development covering Kotlin, Jetpack Compose, Coroutines, Flow, and modern architecture patterns (MVVM, MVI).
angular-proExpert Angular development covering Angular 18+, Standalone Components, Signals, RxJS, and enterprise-grade architecture.
api-design-pro\"Professional API design guidance: contract-first REST/RPC/event interfaces, resource and action modeling, mutation semantics and concurrency (PUT/PATCH/DELETE, upsert, soft delete, bulk), workflow and state-transition APIs, query/filter/search/projection, async jobs and webhooks, consistency and c
api-security-proExpert-level API security based on OWASP API Top 10, including OAuth2, OIDC, Rate Limiting, and JWT hardening.
auth-pro\"Professional authentication and authorization guidance across web/mobile/API systems: identity flows, credential and token lifecycle, session/JWT/OAuth/OIDC/API-key patterns, and policy enforcement models (RBAC/ABAC/ReBAC).\"\nUse this skill when the user asks to design or review auth architecture
aws-proExpert AWS architecture and management covering core services (EC2, S3, RDS, Lambda), IAM security, and Infrastructure as Code (Terraform/CDK).
azure-storage\"Azure Storage Services including Blob Storage, File Shares, Queue Storage, Table Storage, and Data Lake. Provides object storage, SMB file shares, async messaging, NoSQL key-value, and big data analytics capabilities. Includes access tiers (hot, cool, archive) and lifecycle management. USE FOR: bl
blockchain-proExpert Blockchain development covering Ethereum (Solidity), DeFi, L2 solutions, Smart Contract security, and Web3 integration.
browser-skillBrowser automation for AI agents using the vercel-labs/agent-browser CLI. Open web pages, inspect the interactive accessibility tree, click, fill, evaluate, capture screenshots, manage sessions, and run as an MCP server. Use when an agent needs to browse, scrape, verify web apps, or automate actions
bug-discovery-pro\"Professional defect discovery and production-oriented debugging: full methodology (observe, reproduce, narrow, hypothesize, validate, handoff fix, regression prevention), bug taxonomy (logic, state, concurrency, data, integration, config, infra, performance, security), hypothesis-driven experiment
bun-cli-pro\"Professional Bun CLI development: Commander.js subcommands, config persistence, API clients, UX polish with ora spinners and chalk output, and binary compilation.\"\nUse this skill when building CLI tools, designing command hierarchies, managing user configuration, parsing arguments, making HTTP r
business-analysis-pro\"Professional business and product analysis: requirements elicitation, SA-style reasoning, prioritization and traceability, explicit decision frameworks (criteria, weights, sensitivity), product outcomes (segments, value proposition, north star), metrics hierarchy and guardrails, structured risk mo
caching-pro\"Professional distributed and multi-layer caching: patterns (cache-aside, read/write-through, write-behind), end-to-end data flows (read, write, invalidate, warm, refresh), formal consistency models (read-after-write, bounded staleness, eventual, session/monotonic hints), write-path ordering and du
ci-cd-pro\"Production-grade CI/CD system design: pipeline control plane vs runner execution, trigger-to-deploy data flow, immutable artifact and digest promotion, GitHub Actions and GitLab CI/CD patterns, secrets/OIDC/IAM boundaries, caching and parallelism economics, deployment strategies (blue-green, canar
clean-architectureProvides implementation patterns for Clean Architecture, Domain-Driven Design (DDD), and Hexagonal Architecture (Ports & Adapters) in NestJS/TypeScript applications. Use when structuring complex backend systems, designing domain layers with entities/value objects/aggregates, implementing ports and a
clean-code-architecture-pro\"Production-grade clean code and Clean Architecture: dependency rule as a system model (inward stability, ports/adapters, dependency vs control flow), bounded contexts and anti-corruption layers, strangler and incremental migration, failure modes (leaky domain, cycles, anemic model, god modules, di
cli-pro\"Production-grade CLI design and implementation: stdin/stdout/stderr and exit code as a runtime contract, TTY vs non-interactive behavior, SIGPIPE/EPIPE and signal handling, argv/subcommand design, strict vs permissive parsing, config/env/flag precedence, machine-readable output (--json) policies,
cloud-native-agent-pro\"Deploy, scale, and operate AI agents on Kubernetes and cloud-native infrastructure: agent CRDs, MCP server deployment, eBPF observability, Wasm edge inference, and agent scaling patterns.\"\nUse this skill when deploying agents to Kubernetes, designing agent infrastructure as code, scaling agent w
cloudflare-proExpert Cloudflare development covering Workers, R2 Storage, D1 Database, KV, Pages, and Edge computing architecture.
code-packaging-pro\"Production-grade code packaging and CI build automation: artifact graph (wheel, sdist, OCI image) from source to registry; reproducible builds, lockfiles vs semver library ranges, pyproject.toml / build backends (hatchling, setuptools), multi-stage Docker and .dockerignore, GitHub Actions job grap
code-proExpert optimization for Claude-native tools (Claude Code CLI, Projects, Artifacts). Focuses on long-context management and model-specific capabilities.
code-review-proPerform thorough security, quality, and architectural code reviews. Check for vulnerabilities, regression risks, data integrity, test coverage, and API consistency. Use before merging changes, finalizing pull requests, or evaluating critical diffs. Triggers: \"review code\", \"code review\", \"audit
content-analysis-pro\"Production-grade multimodal content analysis: explicit analysis pipeline (modality → decode → segment → extract → verify → report), evidence and provenance rules (page, time, region anchors), grounded vs inferred claims, failure modes (OCR error, sampling gaps, chart number invention, deepfakes, t
cpp-proExpert C++ development covering modern C++20/23 standards, RAII, Smart Pointers, Template Metaprogramming, and performance optimization.
data-analysis-pro\"Production-grade tabular data analysis in Python: analytical pipeline and grain model (row unit, joins, lineage), EDA and cleaning discipline, descriptive statistics and honest visualization, pivot-style workflows, Parquet/CSV/SQLite IO, failure modes (join explosions, timezone bugs, lookahead lea
data-engineering-proExpert Data Engineering development covering ETL/ELT pipelines, distributed processing (Spark, Flink), message queues (Kafka), and data warehouse architecture (Snowflake, BigQuery).
data-science-proExpert Data Science development covering statistical analysis, Exploratory Data Analysis (EDA), machine learning (Scikit-Learn), and data visualization.
debugging-investigationInvestigate complex bugs and unknown failure modes using hypothesis-driven diagnosis. Isolate root causes, inspect execution traces, reproduce failures deterministically, and verify fixes. Use when encountering failing tests, unexpected crashes, slow responses, or elusive runtime errors. Triggers: \
deploy-workflow\"VPS deployment, monitoring, and recovery for Personal KB + Skill Hub at dev.truongsoftware.com: Docker Compose, Caddy SSL, health checks, backups, troubleshooting.\"\nUse this skill when deploying to the VPS, checking production health, restarting services, SSL issues, or backup/restore on ••••.\n
deployment-pro\"Production-grade software deployment and release engineering: deployment runtime model (control plane vs workloads, artifact promotion graph, drift, consistency expectations per platform), CI/CD flows and gates, hosting trade-offs (static, PaaS, containers, serverless, VM), release strategies (rol
design-system-pro\"Production-grade product UI design: design systems, visual foundations, accessibility baselines, data-dense interfaces, guidelines and consistency, web vs native vs desktop patterns, themes and visual direction, token pipeline (primitive → semantic → component → code), failure modes (token drift,
devrel-proExpert Developer Relations (DevRel) covering community building, Developer Experience (DX), advocacy, and technical content strategy.
discussing-proProduction-grade goal discussion and solution exploration: clarifying vague requests into concrete engineering objectives with explicit scope, constraints, success criteria, scored option comparison, tradeoff analysis, and recommendation — plus structured ideation with Socratic questioning, divergen
django-proExpert Django development covering the ORM, Admin interface, Django REST Framework (DRF), and enterprise application patterns.
docker-compose-pro\"Multi-service Docker Compose orchestration: service definitions, networking, volumes, health checks, depends_on, env/secrets, and dev/prod profiles.\"\nUse this skill when writing or reviewing docker-compose.yml, configuring inter-service networking, named volumes, health checks, dependency orderi
docker-pro\"Production-grade Docker and container development: Dockerfile authoring, multi-stage builds, image optimization, Docker Compose, networking, volumes, secrets, BuildKit, and production hardening — plus system model (build context → layers → image digest → runtime), failure modes (cache bust, secret
elasticsearch-proExpert Elasticsearch development covering indexing strategies, DSL queries, Aggregations, and cluster performance tuning.
electron-pro\"Production-grade Electron desktop apps: main vs renderer, preload and contextBridge, secure IPC, trust boundaries, packaging, auto-update, and cross-platform edge cases — plus system model (processes, IPC as API surface), failure modes (XSS→IPC abuse, ABI mismatch, update/partials, signing trust),
engineering-management-proExpert Engineering Management covering SDLC leadership, team building, mentorship, technical strategy, and Agile operations.
executing-pro\"Production-grade plan execution: executing approved plans step by step with dependency-aware task sequencing, checkpoint verification, adaptive replanning, progress tracking, and completion verification — plus execution cycle (execute → checkpoint → adapt → close), failure modes (ignoring dependen
fastapi-proExpert FastAPI development covering async API design, Pydantic validation, Dependency Injection, and high-performance patterns.
feedback-pro\"Production-grade deep feedback for engineering and product: evidence-based review framing, review lifecycle (artifact → finding → action → verification), issue severity calibration, actionable recommendations, communication tone, decision trade-offs (depth vs velocity, async vs sync, block vs foll
figma-mcp-proFigma design-to-code workflow with MCP server integration. Pixel-perfect implementation from Figma designs using structured workflow. Design system component discovery, variable extraction, and incremental screen building.\nUse when implementing Figma designs in code, connecting design system compon
financial-analysis-proUse this skill whenever the user wants to perform financial analysis, financial modeling, valuation, investment analysis, or work with financial data. This includes analyzing financial statements, calculating financial ratios, building financial models, performing company valuations, analyzing inves
fintech-integration-proUse this skill whenever the user wants to integrate with fintech services, payment processing, banking APIs, financial data providers, or work with financial technology systems. This includes payment gateway integration (Stripe, PayPal, Square), banking API integration (Plaid, Yodlee), financial dat
flutter-pro\"Production-grade Flutter development: widgets (Element/RenderObject pipeline), Material 3 / UX, performance and frame budgets, multi-platform edge cases (web, desktop, channels), Dart patterns — plus framework/render pipeline model, failure modes (async gaps, overflow, jank, plugin skew, golden fl
frontend-design-proAnti-slop frontend design with distinctive aesthetics, motion, and spatial composition. Breaks AI-generated clichés (no Inter/Roboto, no purple-on-white) for memorable, professional interfaces with character and purpose.\nUse when building landing pages, marketing sites, portfolios, dashboards, or a
frontend-patternsFrontend development patterns for React, Next.js, state management, performance optimization, and UI best practices.
fullstack-pro\"Comprehensive Fullstack development in ••••: Next.js 15, Server Actions, shadcn/ui, and AI integration.\\nUse this skill to architect, implement, and review end-to-end fullstack features.\"\n\n
fullstack-rag-proBuilding production-grade Retrieval-Augmented Generation (RAG) pipelines with Vector DBs and hybrid search.
game-dev-proExpert game development covering Unity (C#), physics, shaders, optimization, and game architecture patterns (ECS, Singleton, State Machine).
gatekeeper-skillEnforce policy, security, and verification gates before dangerous operations or transitions. Evaluates evidence against rules and explicitly allows or blocks actions. Use before deployment, publishing, irreversible mutations, or merging. Triggers: \"gate\", \"check gate\", \"gatekeeper\", \"allow or
gemini-api-devUse this skill when building applications with Gemini models, Gemini API, working with multimodal content (text, images, audio, video), implementing function calling, using structured outputs, or needing current model specifications. Covers SDK usage (google-genai for Python, @google/genai for JavaS
git-operations-pro\"Production-grade Git workflow: conventional commits, branch strategies, merge vs rebase vs squash discipline, PR hygiene, inspection (log, blame, bisect), recovery (revert vs reset) — plus repository/history/ref model (DAG, three trees), failure modes (force-push, leaked secrets, submodule drift,
git-worktree-proIsolate git workspace and branch operations using git worktrees. Check repository status, avoid dirty commits, and isolate feature or fix branches cleanly without switching checkout directories. Use when starting non-trivial tasks or isolating risky changes. Triggers: \"git worktree\", \"isolate bra
go-proExpert Go (Golang) development covering concurrency, interface-driven design, testing, and cloud-native application patterns.
graphql-pro\"Production-grade GraphQL: schema design, resolver architecture, execution model (parse → validate → execute), performance (N+1, DataLoader, complexity limits), security (field-level authZ, introspection, query abuse), federation/monolith trade-offs — plus failure modes (null propagation, DoS queri
grill-me-proConduct a relentless, systematic interview about a plan or design — one question at a time — until every branch of the decision tree is resolved and both parties share a clear understanding. Does NOT produce a plan; produces shared understanding. Use when the user wants their design stress-tested, t
image-processing-pro\"Production-grade raster image processing with Pillow (PIL): open/save, resize, crop, rotate, format conversion (JPEG, PNG, WebP), compositing, thumbnails, EXIF orientation, batch workflows — plus pixel pipeline model (decode → mode → encode), failure modes (decompression bomb, EXIF leak, CMYK drif
infrastructure-as-code-pro\"Infrastructure-as-Code: Terraform and Pulumi module design, state management, workspace strategy, drift detection, policy-as-code, and testing (Terratest, OPA). Covers AWS/GCP/Azure provider patterns, remote state backends (S3+DynamoDB, GCS, Terraform Cloud), secret injection, and CI/CD pipeline i
ios-proExpert iOS development covering Swift, SwiftUI, Swift Concurrency, Combine, and native architecture patterns (MVVM, Composable Architecture).
java-proExpert Java development covering modern Java 21+, Virtual Threads (Loom), JVM tuning, and enterprise architecture patterns.
javascript-pro\"Production-grade JavaScript: architecture and maintainability, runtime semantics (event loop, microtasks, ESM/CJS graph), async flows, performance — plus execution/module model, failure modes (lost Promises, `this`, pollution, FP dates, dual-package), trade-offs (JS vs TypeScript, mutability), qua
karpathy-coding-pro\"Andrej Karpathy's 4 principles for LLM-assisted coding: Think Before Coding, Simplicity First, Surgical Changes, Goal-Driven Execution. Derived from Karpathy's observations on common LLM coding pitfalls.\"\nUse this skill when the user wants to improve LLM coding behavior, reduce unnecessary chang
kb-workflowMemory and Knowledge Base lifecycle for agents: search before work, write structured solutions after solving problems, choose the right storage tier. Default memory is Markdown files in .aix/runtime/memory/ — always available, no install required. Optional: start the KB service (SQLite/FTS5 knowledg
kubernetes-proExpert Kubernetes (K8s) orchestration covering manifests, Helm, Kustomize, security, networking, and cluster management.
machine-learning-proExpert Machine Learning development covering Deep Learning, PyTorch/TensorFlow, Model Fine-tuning, NLP, and Computer Vision.
mapping-codebaseRapidly explore, index, and orient within unfamiliar codebases. Trace data flows, dependency graphs, architecture patterns, and key entry points without reading entire repositories. Use when starting work on new projects or navigating unknown subsystems. Triggers: \"map codebase\", \"orient in repo\
market-research-promcp-server-promicroservices-promlops-promobile-design-promongodb-promotion-design-pronestjs-neo4j-pronestjs-pronetwork-infra-pronextjs-15-pronextjs-pronextjs-security-scanocr-proparallel-agents-propdf-properformance-tuning-proplanning-proplatform-design-propostgres-patternspostgresql-proprisma-postgresproduct-management-proprompt-engineering-propython-proreact-native-proreact-proredis-proremember-prorepo-tooling-proreport-writerrouter-prorust-prosecurity-prosecurity-reviewself-improve-agent-prosenior-architectsenior-backendsenior-frontendsenior-securityseo-proshadcn-mastery-proskill-authoringskill-creator-proskills-self-review-prosolidity-securityspring-boot-prosql-data-access-prostrategic-consulting-prostream-rtc-prosustainable-design-prosync-custom-to-reposystem-design-overviewsystem-designsystematic-debugging-protauri-protechnical-writing-protest-driven-development-protesting-proto-issues-proto-prd-protool-discovery-skillttd-protypescript-proui-design-brain-proui-reverse-engineer-proui-stack-proui-ux-system-prousing-aixusing-harnessux-design-provercel-deployment-proverificationverify-provibe-coding-provps-devops-provue-proweb-research-prowebsocket-prowriting-skillsskill"${CLAUDE_PLUGIN_ROOT}/hooks/session-start.sh"
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Package any website or local web build into a lightweight desktop app with Pake CLI.
1 skill
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.