Agents and harnesses

Design and operate systems that combine model decisions with tools.

• Agent Design Patterns (510)
• Harnesses And Frameworks (4)
• Tool Use And Mcp (36)
• Coding Agents (2)
• Computer And Browser Use (1)
• Multi Agent Systems (2)

555 resources · page 9 of 18

• Amp is now available. Here's how I use it.

  Creator not supplied; see original source.

  Thorsten Ball's notes on how he uses Amp, the coding agent he works on, in daily development.

  Intermediate · unknown · unknown resource · en

• How to Build an Agent (Amp)

  Creator not supplied; see original source.

  Amp post by Thorsten Ball showing how to build a code-editing agent in under 400 lines of Go.

  Intermediate · unknown · unknown resource · en

• Building Effective AI Agents

  by Anthropic

  Workflows vs agents and the core patterns; argues to start simple. Readable without code.

  Intermediate · West · closed resource · en

• Best practices for Claude Code

  by Anthropic

  Concrete coding-agent workflow habits: plan, verify, manage context.

  Intermediate · West · closed resource · en

• Claude prompt caching docs

  Creator not supplied; see original source.

  Anthropic documentation on prompt caching: caching prompt prefixes to cut cost and latency, with cache lifetimes.

  Intermediate · unknown · unknown resource · en

• Claude remote MCP servers

  Creator not supplied; see original source.

  Anthropic documentation listing remote MCP servers that can be connected to Claude through the MCP connector API.

  Intermediate–Advanced · unknown · unknown resource · en

• Transkribus: Automatically transcribing your documents

  Creator not supplied; see original source.

  Transkribus help article on automatically transcribing documents with text recognition models.

  Intermediate–Advanced · unknown · unknown resource · en

• Atlassian Rovo

  Creator not supplied; see original source.

  Atlassian Rovo product page: AI search, chat and agents across Jira, Confluence and connected work tools.

  Intermediate–Advanced · unknown · unknown resource · en

• Augment blog

  Creator not supplied; see original source.

  Augment Code blog with product updates and engineering posts about its AI coding assistant.

  Intermediate · unknown · unknown resource · en

• A good AGENTS.md is a model upgrade (Augment Code)

  Creator not supplied; see original source.

  Augment Code post measuring how AGENTS.md instruction files affect coding-agent output, with guidance on writing them.

  Intermediate · unknown · unknown resource · en

• What is loop engineering? (Augment Code)

  Creator not supplied; see original source.

  Augment Code post defining loop engineering, the design of agentic software-development loops from trigger to checked outcome.

  Intermediate · unknown · unknown resource · en

• AutoGen documentation

  Creator not supplied; see original source.

  AutoGen documentation: Microsoft's multi-agent conversation framework, now in maintenance mode.

  Intermediate · unknown · unknown resource · en

• Axolotl docs

  Creator not supplied; see original source.

  Axolotl documentation: open-source, config-driven tool for fine-tuning and post-training LLMs.

  Intermediate · unknown · unknown resource · en

• Berkeley Agentic AI (F25)

  Creator not supplied; see original source.

  Berkeley CS294/194-196 Agentic AI (Fall 2025) course site with lecture schedule and the AgentX competition.

  Intermediate–Advanced · unknown · unknown resource · en

• Berkeley CS294 LLM Agents (F24)

  Creator not supplied; see original source.

  Berkeley CS294/194-196 Large Language Model Agents (Fall 2024) course site with lectures and readings.

  Intermediate · unknown · unknown resource · en

• Berkeley RDI education hub

  Creator not supplied; see original source.

  Berkeley RDI education hub indexing its LLM Agents, Advanced LLM Agents and Agentic AI courses.

  Intermediate · unknown · unknown resource · en

• Bolt: Prompt effectively

  Creator not supplied; see original source.

  Bolt support guide on prompting effectively: planning first, prompt enhancement and troubleshooting.

  Intermediate · unknown · unknown resource · en

• Bolt Help Center

  Creator not supplied; see original source.

  Bolt Help Center: guides for building and publishing websites and apps with the Bolt AI builder.

  Intermediate–Advanced · unknown · unknown resource · en

• Braintrust blog

  Creator not supplied; see original source.

  Braintrust blog on AI evaluation, LLM observability and eval-driven development of AI products.

  Intermediate · unknown · unknown resource · en

• Agentic Browser Security: Indirect Prompt Injection in Perplexity Comet (Brave)

  Creator not supplied; see original source.

  Brave post showing indirect prompt injection in Perplexity's Comet browser that exfiltrated email and one-time passwords.

  Intermediate · unknown · unknown resource · en

• Brave Search API docs

  Creator not supplied; see original source.

  Brave Search API documentation for querying Brave's independent web search index from applications.

  Intermediate–Advanced · unknown · unknown resource · en

• Browserbase blog

  Creator not supplied; see original source.

  Browserbase blog on headless browser infrastructure for building web-browsing agents.

  Intermediate · unknown · unknown resource · en

• Building effective agents

  Creator not supplied; see original source.

  Anthropic post on building effective agents: workflow patterns versus autonomous agents and keeping designs simple.

  Intermediate · unknown · unknown resource · en

• About Canva MCP (Canva Developers)

  Creator not supplied; see original source.

  Canva MCP documentation: connecting AI assistants to Canva design capabilities through the Model Context Protocol.

  Intermediate–Advanced · unknown · unknown resource · en

• Best practices for Canva MCP

  Creator not supplied; see original source.

  Canva MCP best practices: building reliable Canva experiences in AI assistant applications.

  Intermediate · unknown · unknown resource · en

• Write better prompts to maximize your AI allowance (Canva Help)

  Creator not supplied; see original source.

  Canva Help Center tips on writing prompts for Canva's AI tools for images, video clips and Canva Code.

  Intermediate · unknown · unknown resource · en

• Canva Developers

  Creator not supplied; see original source.

  Canva Developers portal with APIs and documentation for building integrations and apps on Canva.

  Intermediate–Advanced · unknown · unknown resource · en

• Canva Developers SDK

  Creator not supplied; see original source.

  Canva Apps SDK documentation for developing apps that run inside the Canva editor.

  Intermediate · unknown · unknown resource · en

• China Law Translate: AI-generated content labeling measures

  Creator not supplied; see original source.

  China Law Translate page with the text and English translation of China's 2025 measures for labelling AI-generated content.

  Intermediate · unknown · unknown resource · en

• Claude Agent SDK overview

  Creator not supplied; see original source.

  Claude Agent SDK documentation: using Claude Code as a Python or TypeScript library with hooks, subagents, skills and plugins.

  Intermediate · unknown · unknown resource · en

• Claude Agent Skills docs

  Creator not supplied; see original source.

  Claude Platform documentation on Agent Skills: packaging instructions, metadata and resources that extend Claude.

  Intermediate · unknown · unknown resource · en

• Best practices for Claude Code

  Creator not supplied; see original source.

  Claude Code best practices: configuring the environment, CLAUDE.md, permissions and running parallel sessions.

  Intermediate · unknown · unknown resource · en

Plain-text Markdown copy