Scheduled AI work
Learn the control loop, then explore repeatable workflows and ready-to-use skills.
Explore automation skills →Kendr Blog
Practical ideas for AI that actually does the work.
Understand the systems behind useful personal AI—from automation, memory, and local models to agent protocols, research workflows, and building software with AI.
Agentic AI systems explained: goals, tools, memory, and control
A plain-language map of how AI moves from answering questions to pursuing real work across multiple steps.
Read the agentic AI guide →Agents and automation
Move from an answer to a completed workflow.
Learn how goals, memory, tools, approvals, and scheduled execution fit together without hiding the tradeoffs.
Agentic AI systems explained
How goals, planning, tools, observations, memory, approvals, and evaluation work as one system.
How teams should design agentic workflows
A practical model for routing, execution lanes, checkpoints, artifacts, and human control.
Agent control protocols explained
Understand admission checks, policy boundaries, approval gates, and safer agent execution.
Models and context
Give AI the right context—not simply more context.
Understand retrieval, reusable memory, evidence, and the context systems that make model output more useful.
Retrieval augmented generation explained
Learn embeddings, chunking, retrieval, reranking, citations, and evaluation without the jargon wall.
Build an AI knowledge base from ongoing work
Preserve useful findings, decisions, artifacts, and context instead of repeatedly rediscovering them.
What makes a deep research workspace useful?
Connect the original question, source trail, reasoning, deliverable, and the knowledge worth reusing.
Builders and protocols
Understand the plumbing behind capable AI products.
Clear explanations of the protocols and packaging systems that let agents use tools, share work, and remain governable.
Model Context Protocol explained
A practical guide to MCP hosts, clients, servers, tools, resources, prompts, and transports.
Agent-to-agent protocol explained
Learn agent discovery, task handoff, messages, artifacts, delegation, and interoperability.
MCP skill packs for repeatable AI tools
Package capabilities so agents can discover, install, understand, and safely reuse them.
Workflows and operations
Apply the technology to work people already recognize.
Concrete guides for engineering, operations, evaluation, and research workflows where AI can create a reviewable result.
Architecture review with an AI agent
Analyze a repository, surface technical risk, and turn findings into a phased engineering plan.
CI/CD failure triage with AI
Use logs, test output, and recent changes to isolate likely causes and prepare a safe fix.
Cloud operations AI assistants
Inspect systems, diagnose issues, preserve evidence, and prepare approval-gated remediation.
Competitive intelligence AI workflows
Turn market signals, competitor claims, and internal observations into decision-ready briefs.
Vendor evaluation AI checklist
Compare products, review risks, preserve evidence, and keep the decision useful at renewal time.
AI agent protocols compared
See where MCP, RAG, A2A, ACP, and agentic AI fit—and where they do not overlap.
Explore by problem
Follow a topic from explanation to implementation.
Each cluster connects foundational concepts to concrete workflows, skills, and product capabilities.
Tools and protocols
Understand MCP and see how connected tools become safe, reusable capabilities.
Explore MCP and skills →Memory and knowledge
Move from raw documents to evidence-backed context that stays useful over time.
Explore AI memory →AI-assisted software
Use architecture review, CI triage, and code-aware workflows to build with context.
Explore building with AI →Turn the ideas into working routines.
Start with Kendr on the web, then install the desktop app to connect local tools, schedule work, use local models, and build with Kendr Code.