AI agent builder playbook: tools, memory, multi-step reasoning
Design AI agents with tool calling, memory management and multi-step reasoning — ChatGPT, Claude, Gemini patterns plus Slack/Telegram/web integrations
- What
- Design AI agents with tool calling, memory management and multi-step reasoning — ChatGPT, Claude, Gemini patterns plus Slack/Telegram/web integrations
- Cost
- Free
- Needs
- a use case for an AI agent (support bot, research assistant, automation); API keys for whichever LLM and third-party services the agent will call (e.g. OpenAI/Anthropic, Slack, Telegram) — keep keys in the secure vault, never in code
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor — @claude-office-skills' playbook for designing and building AI agents with tools, memory and multi-step reasoning: agent architecture (input → agent → output with tools, memory and knowledge layers), an agent-types table (reactive, conversational, tool-using, reasoning, multi-agent with complexity ratings), tool-calling patterns (tool definition YAML with parameters schema, common tool categories — data retrieval, actions, computation, generation — and an n8n agent workflow example), memory patterns (buffer, summary, vector and entity memory with implementations, plus context-window management strategies — sliding window, relevance-based, hierarchical — with a token budget), multi-step reasoning (ReAct thought/action/observation loops, a planning agent with plan → execute → synthesize steps), platform integrations (Slack bot with thread context, Telegram bot with voice/image/document handlers, web chat frontend/backend with JWT sessions), and ready-to-adapt agent templates (customer support agent with escalation triggers, research agent with sources-cited output), all illustrated with a worked "appointment scheduling agent" example. Works in English and Chinese; recommended models claude-opus-4/claude-sonnet-4. MIT-licensed. Skill Harbor never reviews the code, review it yourself before use. Discovered via skills.sh.
Version:
Install
Prerequisites: a use case for an AI agent (support bot, research assistant, automation); API keys for whichever LLM and third-party services the agent will call (e.g. OpenAI/Anthropic, Slack, Telegram) — keep keys in the secure vault, never in code Install "AI agent builder playbook: tools, memory, multi-step reasoning" for me. It gives my agent @claude-office-skills' AI-agent design guide: agent architecture and types, tool-calling patterns (definitions, categories, n8n integration), memory patterns (buffer/summary/vector/entity, context-window strategies with token budgets), multi-step reasoning (ReAct, planning agent), platform integrations (Slack, Telegram, web chat), and ready templates (customer support, research) with a worked appointment-scheduling example. Works in English and Chinese. MIT-licensed. Repository: https://github.com/claude-office-skills/skills/blob/main/ai-agent-builder/SKILL.md 1. Fetch the SKILL.md file (and any helper files) from the repository path into a temporary folder and summarize what it does in one or two sentences. 2. Safety check: review the SKILL.md and scripts for anything suspicious (unexpected network calls, shell commands, credential harvesting). This repo should contain zero secrets in code, credentials only via the secure vault, allowed hosts declared in the SKILL.md. Verify that holds here; STOP on any red flag and tell me. 3. Install it as a skill: copy SKILL.md and its helper files into the agent's skills directory, in a folder named "ai-agent-builder". 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (e.g. pick the target agent use case and save my API keys in the vault; then tell the agent which template to start from). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.