# WorkerKit blog > How AI workers are built, run and paid for: kits versus plugins and skills, managed runtimes, schedules and triggers, model choice and what a run costs. Posts listed: 38. Newest first. This file is an index. Each post page carries its own full text in the page's structured data, so a post can be read without executing JavaScript. ## Posts - [Cost per acceptable AI result](https://workerkit.ai/blog/cost-per-acceptable-ai-result): Measure AI worker value using accepted outcomes, including execution, retries, human review, correction and maintenance in the total cost. By Adam Shamos. Published: 2026-10-29. Updated: 2026-10-29. Tags: evaluation, costs, workforce. - [From AI tools to an AI workforce](https://workerkit.ai/blog/from-ai-tools-to-an-ai-workforce): How people, AI models and workers operate together, with managed cloud infrastructure, shared knowledge and company control over each job. By Adam Shamos. Published: 2026-10-29. Updated: 2026-11-13. Tags: workforce, adoption, operations. - [Claude Code MCP Tasks for long-running jobs](https://workerkit.ai/blog/claude-code-mcp-tasks): Claude Code now runs long MCP tool calls as tasks. The server returns a handle and Claude keeps working. How it works, its limits, and where workers fit. Published: 2026-10-23. Updated: 2026-10-23. Tags: claude-code, claude, mcp, anthropic, developers. - [AI agents vs RPA: choose by process step](https://workerkit.ai/blog/ai-agents-vs-rpa): Compare AI agents, RPA and workflow automation by the work each step needs. Design a support process with clear handoffs, review and execution records. By Adam Shamos. Published: 2026-10-22. Updated: 2026-10-22. Tags: enterprise, automation, agents, rpa. - [Enterprise AI automation: a practical blueprint](https://workerkit.ai/blog/enterprise-ai-automation): Build enterprise AI automation around defined jobs, app permissions and measurable results, with practical use cases and a process brief for your team. By Adam Shamos. Published: 2026-10-22. Updated: 2026-10-22. Tags: enterprise, automation, workforce, operations. - [Evaluating enterprise AI automation platforms](https://workerkit.ai/blog/enterprise-ai-automation-platform-evaluation): Compare enterprise AI automation platforms on one real job, with mandatory controls, acceptance tests, total operating cost and a clear decision record. By Adam Shamos. Published: 2026-10-22. Updated: 2026-10-22. Tags: enterprise, evaluation, automation. - [What is Claude Duet? Features, access and limits](https://workerkit.ai/blog/claude-duet-explained): Claude Duet is Anthropic's proactive agent app. What it does, who can get it, how usage limits work, what it asks before acting, and where AI workers fit. Published: 2026-10-21. Updated: 2026-10-21. Tags: claude-duet, claude, agents, anthropic, orchestration. - [Claude Duet vs OpenAI Dots vs Meta Muse](https://workerkit.ai/blog/claude-duet-vs-dots-vs-muse): Compare Claude Duet, OpenAI Dots and Meta Muse from their launch material: computer, channels, work while away, approvals, access, cost and connections. Published: 2026-10-21. Updated: 2026-10-21. Tags: claude-duet, openai-dots, meta-muse, agents. - [Hark Pro vs Meta Muse vs Instinct vs Grok Bot](https://workerkit.ai/blog/hark-vs-muse-vs-instinct-vs-grok-bot): Compare Hark Pro, Meta Muse, Instinct and Grok Bot from their own launch pages and terms: computers, memory, approvals, pricing and connections. Published: 2026-10-06. Updated: 2026-10-06. Tags: hark, meta-muse, instinct, grok-bot, agents. - [OpenAI Decisions API: the public beta explained](https://workerkit.ai/blog/openai-decisions-api-public-beta): OpenAI's Decisions API is in public beta: typed questions on gpt-6-luna at $0.10 per 1M input tokens. How it works, what it costs, and how it compares with Jev. Published: 2026-10-06. Updated: 2026-10-08. Tags: openai-decisions-api, jev, decision-models, pricing. - [OpenAI Dots vs Meta Muse vs Grok Bot](https://workerkit.ai/blog/dots-vs-muse-vs-grok-bot): Compare OpenAI Dots, Meta Muse and Grok Bot using their launch announcements: computers, channels, connected apps, access and approvals. Published: 2026-09-29. Updated: 2026-10-21. Tags: openai-dots, meta-muse, grok-bot, agents. - [OpenAI Decisions API vs Jev, field by field](https://workerkit.ai/blog/openai-decisions-api-vs-jev): OpenAI Decisions API vs TypeSafe Jev, now that both contracts are public. Question types, confidence, refusals, images, limits, data terms and a 2.4x price gap. Published: 2026-09-29. Updated: 2026-10-06. Tags: openai-decisions-api, jev, decision-models, luna. - [What are OpenAI Dots? Features and access](https://workerkit.ai/blog/openai-dots-explained): OpenAI Dots are always-on agents in ChatGPT. Learn how to get a dot, what it can do, launch pricing and limits, and how its approvals work. Published: 2026-09-29. Updated: 2026-10-21. Tags: openai-dots, agents, chatgpt, orchestration. - [Decision model vs agent: the two halves](https://workerkit.ai/blog/decision-models-vs-agents): Almost every job handed to an AI agent is two jobs stapled together: a judging half and a working half. They want different machinery. Published: 2026-09-16. Updated: 2026-09-16. Tags: decisions, agents, cost. - [Jev: TypeSafe's System One model explained](https://workerkit.ai/blog/jev-and-decision-models): Jev is TypeSafe's System One model. It answers typed questions with probabilities and writes no text at all. What it does, costs, and cannot do. Published: 2026-09-16. Updated: 2026-10-06. Tags: jev, typesafe, decision-models, models. - [Screen cheap, spend expensive: agent triage](https://workerkit.ai/blog/screen-cheap-spend-expensive): A worker that reads 400 items to act on 9 bills you for reading 400. Three ways to put a cheap screening pass in front of the expensive work. Published: 2026-09-16. Updated: 2026-09-16. Tags: pricing, models, triage. - [Connecting Meta's Muse Code to your AI workers](https://workerkit.ai/blog/muse-code-and-ai-workers): Meta's Muse Code shipped with MCP support. Point it at the WorkerKit MCP server and the coding agent can read, run and schedule your AI workers. Published: 2026-08-31. Updated: 2026-10-08. Tags: muse-code, mcp, orchestration, developers. - [Muse Code connectors: adding MCP servers](https://workerkit.ai/blog/muse-code-connectors): How connectors work in Meta's Muse Code: the mcp_servers settings format, both transports, the sandbox caveat, and a first server worth adding. Published: 2026-08-31. Updated: 2026-10-12. Tags: muse-code, mcp, connectors, developers. - [A managed runtime for unattended AI jobs](https://workerkit.ai/blog/managed-runtime-for-unattended-ai-jobs): An unattended AI worker is not a model deployment, it is a job deployment. What a managed runtime standardizes is everything that job needs to keep running. By Adam Shamos. Published: 2026-08-29. Updated: 2026-08-29. Tags: runtime, platform, reliability. - [Bot directory: what to check before installing](https://workerkit.ai/blog/bot-directory): A bot directory is a browsable catalog of bots you can install. Here is what a listing has to tell you, and the five checks worth making before you run one. Published: 2026-08-23. Updated: 2026-10-08. Tags: directory, bots, kits, concepts. - [Orchestrating an SEO agentic workflow](https://workerkit.ai/blog/orchestrating-an-seo-agentic-workflow): One worker finds what deserves a page, one drafts it, one publishes it. How to chain three AI workers into an SEO pipeline, and where the human gate belongs. Published: 2026-08-17. Updated: 2026-08-17. Tags: seo, orchestration, content, workers. - [AI agent vs AI worker: where the line is](https://workerkit.ai/blog/ai-agent-vs-ai-worker): An AI worker is an agent with a persistent job, permissions, limits and run record. Compare the terms without confusing scheduled work with ownership. Published: 2026-08-14. Updated: 2026-10-29. Tags: workers, agents, concepts. - [Case study: a worker that texts real people](https://workerkit.ai/blog/case-study-appointment-reminder-texter): Appointment Reminder Texter is the shape where a mistake reaches a stranger. What contact rules actually do, and why this one earns a narrow send stage. Published: 2026-08-13. Updated: 2026-08-13. Tags: case-study, outbound, contact-rules. - [Case study: a morning brief that reads only](https://workerkit.ai/blog/case-study-executive-daily-briefing): A walkthrough of Executive Daily Briefing, the read-only shape: three apps, no write access anywhere, and why that makes it the right first worker. Published: 2026-08-12. Updated: 2026-08-12. Tags: case-study, briefing, scheduling. - [What an unattended AI fleet needs to fail well](https://workerkit.ai/blog/fail-safely-unattended-fleet): Unattended agents fail plausibly rather than loudly. Five design choices decide whether a bad run costs an hour or a quarter, and none is model quality. Published: 2026-08-11. Updated: 2026-08-11. Tags: reliability, runtime, safety. - [How to pick a model for an AI worker](https://workerkit.ai/blog/how-to-pick-a-model-per-job): Benchmarks measure somebody else's task. The only comparison that counts is the same run, on your instruction and your data, judged by you. Published: 2026-08-10. Updated: 2026-08-10. Tags: models, evaluation, runtime. - [How to publish a kit others can run](https://workerkit.ai/blog/how-to-publish-a-kit): A worker that works for you is not yet a kit. What has to be true before publishing, and what the permanent address commits you to afterwards. Published: 2026-08-07. Updated: 2026-10-08. Tags: publishing, mcp, kits. - [Why an instruction is not a safety control](https://workerkit.ai/blog/instruction-is-not-a-control): Telling a model what not to do is a request it weighs against everything else. At unattended scale, a rule that holds 99.5% of the time fails every month. Published: 2026-08-06. Updated: 2026-08-06. Tags: safety, access, concepts. - [How to roll out an AI worker safely](https://workerkit.ai/blog/rolling-out-an-ai-worker-safely): Four stages from read-only to autonomous, what to check before each promotion, and the signals that mean go back a step rather than push on. Published: 2026-08-05. Updated: 2026-08-05. Tags: rollout, evaluation, safety. - [Schedule or webhook? How an AI agent starts](https://workerkit.ai/blog/schedule-or-webhook-trigger): Polling burns capacity discovering nothing happened. Triggers fail silently. The right answer for most unattended workers is both, and the split is not obvious. Published: 2026-08-04. Updated: 2026-08-04. Tags: schedules, triggers, runtime. - [What to automate first, and what not to](https://workerkit.ai/blog/what-to-automate-first): Most teams pick their worst job to automate first. Three tests that predict whether a job will work unattended, and the ones that reliably fail. Published: 2026-08-03. Updated: 2026-08-03. Tags: job-selection, getting-started, workers. - [When a kit is 80% of what you need](https://workerkit.ai/blog/when-a-kit-is-eighty-percent-right): The kit does most of the job and gets one thing wrong. What you can change after installing, what you cannot, and when to start from scratch instead. Published: 2026-07-31. Updated: 2026-10-08. Tags: kits, customisation, getting-started. - [When a worker writes to your CRM](https://workerkit.ai/blog/workers-and-systems-of-record): A wrong CRM field looks recoverable and is not. What changes when an AI worker writes into a system other people read and act on later. Published: 2026-07-30. Updated: 2026-07-30. Tags: crm, data-quality, workers. - [How to write an instruction that holds](https://workerkit.ai/blog/writing-an-instruction-that-holds): A prompt that works in testing drifts by run four hundred. What separates a chat prompt from a standing instruction an unattended worker can run on. Published: 2026-07-29. Updated: 2026-07-29. Tags: instructions, craft, workers. - [Inbound lead qualification with AI workers](https://workerkit.ai/blog/case-study-inbound-lead-qualifier): How Inbound Lead Qualifier reads an inquiry, updates the CRM and drafts a reply, with event triggers, review boundaries and cost planning. Published: 2026-07-28. Updated: 2026-10-29. Tags: case-study, sales, triggers. - [Case study: a support inbox triage worker](https://workerkit.ai/blog/case-study-support-inbox-triager): A walkthrough of deploying the Support Inbox Triager kit: the access it needs, the schedule, the safety rules, what it costs and where it should stop. Published: 2026-07-27. Updated: 2026-10-29. Tags: case-study, support, triage. - [Agent skills vs plugins vs MCP vs kits](https://workerkit.ai/blog/plugins-skills-and-kits): Skills teach a technique, plugins package it for a client, MCP connects your systems, and kits are workers that run without you. Where each boundary sits. Published: 2026-07-24. Updated: 2026-10-08. Tags: kits, skills, plugins, mcp. - [What an AI worker actually costs to run](https://workerkit.ai/blog/what-an-ai-worker-costs): AI worker pricing runs on three meters: the plan, the tool calls and the model tokens. What each measures on WorkerKit, with real figures and worked examples. Published: 2026-07-23. Updated: 2026-10-29. Tags: pricing, runtime, models. ## Notes for AI agents - The rendered blog is at https://workerkit.ai/blog. - Definitions of the vocabulary these posts use: https://workerkit.ai/glossary.md. - Product overview: https://workerkit.ai/about.md. Pricing: https://workerkit.ai/pricing.md. Every public kit: https://workerkit.ai/kits.md. Crawler index: https://workerkit.ai/llms.txt.