Reference Hub
Guides and reference material for AI platforms, tool comparisons, and the technologies we use to build and deliver solutions.
Tool & Platform References
AI Platforms
What each platform offers — products, pricing, and where they fit
Anthropic
Claude chat, Claude Code, Cowork, Claude in Chrome, Claude in Excel, the API and Console, MCP, and enterprise features
Read guide →OpenAI
ChatGPT plans, GPT and o-series models, Codex, DALL-E, Whisper, custom GPTs, and the API platform
Read guide →Google AI
Gemini, Workspace AI, AI Studio, Vertex AI, NotebookLM, AI Overviews, Gemma — the full product suite
Read guide →Microsoft AI
Microsoft Copilot, M365 Copilot, GitHub Copilot, Azure OpenAI, Microsoft Foundry, Copilot Studio — the full product suite
Read guide →Salesforce AI
Agentforce, Einstein Copilot, Atlas Reasoning Engine, Data 360, Prompt Builder, Model Builder, Trust Layer, and Flow with AI
Read guide →ServiceNow
Now Assist, AI Agents, AI Agent Studio, AI Control Tower, Workflow Data Fabric, Context Engine, EmployeeWorks, and the April 2026 Foundation/Advanced/Prime tiers
Read guide →Open-Source Model Infrastructure
Open-weight models, model hubs, and self-hosted AI — what is available to build on
Meta AI
Llama 3 and 4, Meta AI assistant across Facebook/Instagram/WhatsApp, AI Studio, the Llama API, and Muse Spark
Read guide →Hugging Face
Model Hub, Datasets, Spaces, Inference Providers, Inference Endpoints, Transformers, AutoTrain, Pro, and Enterprise Hub — the full product suite
Read guide →Mistral and DeepSeek
The two leading non-US open-weight providers: Mistral Large 3, Codestral, Pixtral, Le Chat vs DeepSeek V4, R1, Coder, and the API platform
Read guide →Local Agents
Open-source local agent frameworks: setup, security, messaging, and enterprise readiness compared
Read guide →AI Tools
Cross-platform comparisons — which tool to use for a specific task
Chat Assistants
Claude vs ChatGPT vs Gemini vs Copilot vs Perplexity vs Grok: pricing, models, context windows, web search, file upload, and apps
Read guide →Code Assistants
Claude Code vs GitHub Copilot vs Cursor vs Windsurf vs Amazon Q vs Cline vs OpenAI Codex CLI vs OpenCode: pricing, models, IDE support, and agent modes
Read guide →AI Agents
Cowork vs Agentforce vs Copilot Studio vs AgentKit vs LangGraph: pricing, models, tool integration, and deployment
Read guide →Claude Code
The terminal-native coding agent — CLI, Plan Mode, skills, subagents, hooks, slash commands, MCP, plugins, and the loop that uses them well.
Read guide →Cowork
Anthropic's agentic desktop app for knowledge work — skills, plugins, connectors, scheduled tasks, artifacts, folder access, and the Claude in Chrome bridge.
Read guide →Claude Code vs Cowork
Picking between Anthropic's two agentic surfaces — when to reach for which, and the strategic insight that they share primitives.
Read guide →Design + Build Workflow
Using Cowork and Claude Code together to design and build apps — the six-phase pattern and the one anti-pattern that wrecks it.
Read guide →Cowork Prompt Library
Twenty ready-to-fire Cowork prompts categorized by capability — copy, paste, watch what happens.
Read guide →Cowork Common Mistakes
Eight mistakes that show up in nearly every new Cowork user's first month — and the fix for each, before you make them.
Read guide →Cowork by Role
Pick the role that's closest to yours and get a personalized first-install plan — one plugin, two or three connectors, one routine.
Read guide →Cowork Artifact Pattern
How to build a Cowork artifact that re-fetches your data, fails gracefully, and earns its slot in the artifact list. Worked example included.
Read guide →AI Search
Perplexity vs Google AI Mode vs ChatGPT Search vs Grok vs Copilot Search: pricing, citation quality, real-time data, and research depth
Read guide →Knowledge Graphs
When you need a graph database, when you just need graph thinking, and how to capture most of the value in Postgres with edge tables, pgvector, and Apache AGE
Read guide →Enterprise AI
Microsoft 365 Copilot vs Salesforce Agentforce vs Gemini for Workspace vs ServiceNow Now Assist vs Slack AI vs Notion AI: pricing, lock-in, supported workflows
Read guide →Building a Website
Choosing the right tools and services for web projects
Database Providers
Neon vs Supabase vs Turso vs Railway: free tiers, inactivity policies, and Vercel compatibility
Read guide →Email Platforms
Mailchimp vs Kit vs Resend: choosing the right platform for subscriber capture and newsletters
Read guide →Website Contact
Calendly vs Cal.com vs Formspree vs custom: choosing how visitors reach you
Read guide →Hosting and Deploy
Vercel vs Netlify vs Cloudflare Pages vs Railway vs Render vs Fly.io: free tiers, build limits, and framework support
Read guide →CSS Frameworks
Tailwind vs Bootstrap vs Bulma vs vanilla CSS vs Pico vs Open Props: bundle size, customization, dark mode, and AI-friendliness
Read guide →Database ORMs
Drizzle vs Prisma 7 vs Kysely vs TypeORM vs Sequelize vs raw SQL: type safety, bundle size, serverless cold starts, and AI-friendliness
Read guide →AI Safety and Security
Operating autonomous agents safely — defenses, threat models, and tool-specific safety guidance
Using Cowork Safely
Anthropic's official Cowork safety guidance — file access, scheduled tasks, 'Act without asking' mode, computer use, MCPs and plugins, cross-app sharing, and mobile-as-remote-control
Read guide →Prompt Injection: Mental Models
The eight mental models that organize every other prompt-injection defense — instructions vs data, the lethal trifecta, direct vs indirect, the agent surface, trust boundaries, capability vs autonomy, defense in depth, and the Stranger Test
Read guide →Prompt Injection: Threat Taxonomy
How injection attacks actually arrive — direct vs indirect, document/web/email/multimodal carriers, tool-result injection, exfil via tool chains, memory poisoning, confused deputy patterns
Read guide →Prompt Injection: Defenses
Defense in depth, applied — input limits, tagged inputs, capability restriction, scoped credentials, human-in-the-loop gates, output validation, provenance, sandboxing, monitoring, incident response
Read guide →Prompt Injection: File-Reading Agents
Tool-specific guidance for Cowork, Desktop Commander, and similar — PDF/DOCX/XLSX payload anatomy, image injection, cross-file chains, working-folder discipline, capability restriction
Read guide →Prompt Injection: MCP Servers
MCPs as dependencies — tool-result injection, tool definition tampering, plugin bundles, chain-of-tool attacks, vetting checklist, supply-chain discipline applied to AI tool servers
Read guide →Prompt Injection: Browser Agents
The web as injection surface — hidden HTML, visible-but-disguised payloads, navigation chains, SEO poisoning, form-fill exfil, Claude in Chrome consent, per-session site allowlists
Read guide →Prompt Injection: Email and Chat Connectors
Anyone with your email or workspace ID can write into your agent — email body payloads, calendar invites, Slack/Teams messages, attachments, reply-as-exfil, sender allowlisting limits
Read guide →Disciplines
The practitioner moves and concept anchors that the tool comparisons sit on top of
Context Engineering
The practice that separates AI deployments that drift from ones that hold up. Curation, metadata, governance, audit.
Read guide →Agentic RAG
The patterns that make managed retrieval-augmented generation reliable enough to ship — curated index, metadata, multi-source, citation, audit — and the vendor landscape behind them.
Read guide →Curation Governance for SMB AI
The one-page rubric pattern. The written artifact that turns context-engineering discipline into something a 25–200 person team can actually adopt.
Read guide →AI Governance Program
The operating system that turns AI policy into a discipline — the integration of artifacts, structure, and material-change response that an auditor can verify. Companion to the Enterprise Governance trail.
Read guide →