


Your business should be running AI agents by now. One agent today. A team by next week. Discord discord.gg/creao-ai



This is the four-step agent build we'd hand anyone on day one. One blog post in, five posts out, on a weekly schedule, approval kept on until the agent earns full auto. If you can describe it, CREAO can build it.
Turning one blog post into a week of content is the job most creators keep putting off. An agent can write the first drafts for you. You only approve them. No code. Here's the setup, step by step: → 1. Pick your source One blog post, a video transcript or a newsletter you already wrote. → 2. Tell it exactly what you want Try: "Read this blog post. Write five posts for this week: one bold take, one how-to list, one story, one question and one quote. Match my tone: short lines, plain words, no hype. Send me the drafts to approve before anything goes out." → 3. Put it on a schedule Weekly, same day. Nothing to remember. → 4. Keep approval switched on Anything that sends waits for your sign-off. Move it to full auto only after a few clean runs. Now the honest part: → The first drafts won't sound exactly like you. Edit them, then tell the agent what you changed. → It works from what you give it. A thin post gives you thin drafts. → One small agent that runs every week beats a content system you never finish. Pick one post you've already published and set this up today. Not after another tutorial. Today. You can build any agent on @CreaoAI if you can describe it, CREAO can build it. Bookmark this so you have the prompt, and follow me for more agent builds you can actually copy.

the full breakdown, with a comparison table, is here 👇 creao.ai/blog/ai-agents-vs-a…

A client hasn't paid. Here's how three kinds of AI tools handle it: Automation: emails the same reminder to everyone, even the client who paid yesterday AI assistant: writes a great reminder, but only if you remember to ask AI agent: sees Sam already paid, skips him, and drafts a firmer note for the client who's ignored you twice Which one is running your business right now? 👇


"Discovering Effective Agent Memory Designs through Program Evolution" — new paper (arXiv preprint, Nov 12, 2026). arxiv.org/abs/2611.17068 In plain terms: agent memory is the code that saves what a user said in past chats and looks it up later to answer a new question. Instead of hand-designing that code, MAPE lets a language model rewrite the whole memory program over and over, starting from one that stores nothing, keeping changes that score better. The winning designs beat hand-built memory systems and reveal two simple patterns you can copy. 3 distinct takeaways for product builders: - Keep the raw chat, not only the summary. Extracted notes (short facts pulled out by a model) drop details. Removing the retrieved raw conversation turns from the best design cut accuracy by 12.17 points on LoCoMo and 6.5 on LongMemEval. Store original messages next to your notes and return both at lookup time. - Prepare the found evidence by question type before answering. Their "Query-Routed Scaffolding" sends each question down a path that fits it: date arithmetic in code for time questions, item listing plus code counting for "how many" questions. Bolted onto five existing memory systems (Mem0, A-Mem, LightMem, SimpleMem, Zep), it raised accuracy by 2.75 to 9.25 points on LongMemEval. Do math in code, not in the model's head. - Score cost, not just accuracy. Searching for accuracy alone kept adding model calls (one step raised cost about 6x for a small gain). Searching for accuracy and cost together moved work from save time to question time: 90% and 40% fewer model calls, and on LoCoMo with Qwen 81.33% accuracy, 12.01 points above SimpleMem at 35% of its serving cost. If each chat is asked few questions, skip heavy processing when saving. Experiment setup (from the paper): - Method: MAPE, evolutionary program search (a model proposes code edits, each is tested, a diagnosis agent reads failure traces). MAPE-A optimizes accuracy only; MAPE-P optimizes accuracy and cost together. - Benchmarks (fixed test sets): LoCoMo and LongMemEval-S, long multi-session conversations with questions about them; 20% used for search, 80% held out for final scores. - Serving models: GPT-4.1 mini and Qwen3-30B-A3B-Instruct-2507; embeddings (text-to-vector for similarity search) from text-embedding-3-small; GPT-5 mini judges answer correctness. - Baselines: full conversation in the prompt, plain RAG (retrieve messages by similarity, then answer), Mem0, A-Mem, LightMem, SimpleMem, Zep. - Metrics: answer accuracy (%) and serving cost in dollars per 1,000 questions, including the cost of building memory.


👇
Release notes are the chore everyone puts off until Friday evening. An agent can write the first draft for you. You only approve it. Here's the setup, step by step: → 1. Connect GitHub In CREAO, give the agent read access to the repo you ship from. → 2. Tell it exactly what you want Try: "Every Friday at 5pm, read this week's merged pull requests. Group them into New, Improved and Fixed. Write the notes in plain language a customer can follow. Send me the draft to approve before anything is published." → 3. Put it on a schedule Weekly, same time. No reminder to set, no chore to remember. → 4. Keep approval switched on Anything that writes or sends pauses for your sign-off. Move it to full auto only when you trust the output. Now the honest part: → The first draft won't be perfect. Read it before you approve it. → It writes what your pull requests say. Vague titles give you vague notes. → One small agent running every week beats ten tools you never open. Pick one repo and set this up today. Not after another tutorial. Today. You can build any agent on @CreaoAI if you can describe it, CREAO can build it. Bookmark this so you have the prompt, and follow me for more agent builds you can actually copy.

Release notes are the chore everyone puts off until Friday evening. An agent can write the first draft for you. You only approve it. Here's the setup, step by step: → 1. Connect GitHub In CREAO, give the agent read access to the repo you ship from. → 2. Tell it exactly what you want Try: "Every Friday at 5pm, read this week's merged pull requests. Group them into New, Improved and Fixed. Write the notes in plain language a customer can follow. Send me the draft to approve before anything is published." → 3. Put it on a schedule Weekly, same time. No reminder to set, no chore to remember. → 4. Keep approval switched on Anything that writes or sends pauses for your sign-off. Move it to full auto only when you trust the output. Now the honest part: → The first draft won't be perfect. Read it before you approve it. → It writes what your pull requests say. Vague titles give you vague notes. → One small agent running every week beats ten tools you never open. Pick one repo and set this up today. Not after another tutorial. Today. You can build any agent on @CreaoAI if you can describe it, CREAO can build it. Bookmark this so you have the prompt, and follow me for more agent builds you can actually copy.

"The RSI Tax: When Recursive Self-Improvement Is Not Worth the Cost" — new paper (arXiv preprint, Nov 10, 2026). arxiv.org/abs/2611.13524 In plain terms: a self-improving agent learns from its own past tasks by rewriting its prompts, code or skill files between jobs (recursive self-improvement, RSI). This paper bills the tokens (the units LLM providers charge by) that the learning step itself burns, the "RSI Tax", and finds a simple capped notes file matches or beats the heavy methods at roughly a third of the total cost. 3 distinct takeaways for product builders: - Cost accounting: the learning bill can be bigger than the work bill. On one terminal task stream, Meta-Harness spent fewer tokens doing the tasks than the simple baseline (184.13M vs 214.37M), but its 578.45M-token RSI Tax pushed its total to 762.57M vs 214.65M, with slightly lower accuracy (77.5% vs 78.7%). Compare self-improvement methods on total tokens, not per-task tokens. - Where the money goes: verification, meaning re-running the agent on check tasks to see if a proposed change really helps, was 74.8–93.6% of the RSI Tax for Meta-Harness, EvoSkill and RRSI. EvoSkill spent 497.45M tokens re-checking old candidates vs 182.26M serving the 89 real tasks. Cap or sample verification runs first; that is where the savings are. - Start simple: Simple Reflection with Notes (SRN) only rewrites one note of at most 2,048 tokens from failed tasks, every four tasks. It had the top accuracy in 3 of 4 settings (tied in the 4th), +4.7 points over a never-learning agent for +3.6% tokens, and used 65.6–71.0% fewer tokens than the heavy methods. Ship a bounded lessons file as your baseline before building a search loop. Experiment setup (from the paper): - Methods (6): Static (never learns), ReasoningBank (stores lessons from past reasoning), EvoSkill (evolves reusable skill files), Meta-Harness (searches over the agent's controller code, the "harness"), RRSI (harness edits penalized for overfitting), and SRN. - Models: DeepSeek V4.1 Flash and GPT-6 Luna; all start from the same KIRA agent. - Task streams: SWE-Chain-Evo (105 positions of chained coding tasks across 7 repository chains) and Terminal-Bench 2.1 (89 command-line tasks). - Protocol: each task is scored before its feedback can be used (no peeking); learners update every four tasks. - Metrics: accuracy (% of tasks passing an automatic checker) and total tokens = task execution + RSI Tax, split into verify / explore / write. - Limits: two models, one task ordering, single runs.



Stop writing please pay me emails. Steal this prompt👇 Every weekday at 9am, check my invoices for anything unpaid and overdue. For each one, draft a short, friendly follow-up. Warm the first time, firmer after 14 days. Mention the invoice number, amount and due date. Don't send anything. Show me the drafts so I can approve them in one click. Paste it into CREAO and turn it into an agent.

