Skip to content

Blog

Coding agent setup

How to set up a coding agent: the loop it runs, skills and subagents, MCP servers and CLIs, and when one general agent beats many custom ones.

  • 14 posts, newest first
  1. A swarm of subagents is not throughput Coordinators that fan out to thousands of subagents measure merged pull requests, which is output. The limit on fan-out is how much a team can check well enough that what lands is what customers need. 1 min readShort post
  2. An agent is a loop, and the loop is what you debug A coding agent perceives, decides, acts and observes, then repeats. Naming which of the four steps failed turns a mystery into a fix. 2 min readShort post
  3. Skills hold what agents need to know, sub-agents do the thinking The first architecture decision in an agent setup is sorting what needs to think from what just needs to know. Skills carry static knowledge, sub-agents do dynamic work, and the split decides the cost. 2 min readShort post
  4. A harness is a list of things the model cannot do yet A harness is not plumbing around a model. Every component in it encodes one assumption: a thing the model cannot be trusted to do alone. Read it that way and the harness becomes a checklist. 2 min readShort post
  5. Agent teams win by moving complexity out of the agent Vercel, Manus and Pi got better results by removing tools from the agent. The orchestration belongs in deterministic factory steps around a small, focused agent. 1 min readShort post
  6. Claude Code can now navigate your code like an IDE Claude Code used to search a codebase with grep and text matching. With Language Server Protocol plugins it gets real code intelligence: references, signatures and types. 1 min readShort post
  7. The best agent architecture is a simple while loop Claude Code and the pi agent show the same thing: a main loop, a few general-purpose tools and careful context management beat elaborate orchestration. 1 min readShort post
  8. Agents should write code to call MCP tools, not call them directly In Anthropic's example, direct MCP tool calls took 150,000 tokens and an agent writing code to call the same tools took 2,000. Code execution is the way to use MCP. 2 min readShort post
  9. For tools you own, build a CLI before an MCP server Every MCP server loads all its tools into context before the agent starts. Keep MCP for external services and give your own tools a CLI, scripts or skills, which cost a fraction of the context. 2 min readShort post
  10. Slash commands, MCP, sub-agents and skills form a hierarchy Teams build the same workflow three ways because they treat Claude Code's four overlapping features as equals. They form layers: prompts at the base, MCP and sub-agents beside each other, skills composing the rest. 2 min readShort post
  11. Shallow agents fail at step 20, deep agents plan and remember The prompt, tool, result loop works for 5 to 15 steps and collapses on real work. Deep agents add explicit planning, delegation to sub-agents, persistent memory and heavy context engineering. 2 min readShort post
  12. One general agent with a skill library beats 47 custom agents Teams cram every procedure into monolithic agent prompts. Agent Skills replace them with one general agent and a composable library that loads knowledge in layers. 2 min readShort post
  13. Use subagents as researchers, not implementers Subagents that implement directly lose context between sessions. Subagents that research, write their plans to the file system and leave the implementation to the main agent work in production. 2 min readShort post
  14. Generic coding agents hit a limit, custom agents go past it Claude Code, Cursor and Gemini CLI are built for the average codebase. Teams that scale with agents move from better agents to more agents to custom agents built for their own domain. 3 min readShort post

Go deeper

Other topics