Series
AI Tooling for Developers
A seven-part guide to the agent ecosystem beyond your coding IDE — MCP vs CLI tradeoffs, Jira/Notion integrations, release note automation, Paperclip, OpenClaw, and Hermes.
7 parts · first published
- 01
MCP vs CLI: The Token Cost You're Not Tracking
Every MCP tool call ships your entire tool schema to the model. On a long session that's thousands of tokens before you've done anything. CLI tools don't have that problem.
· 5 min read - 02
Setting Up the Jira MCP Server (And When Not To)
Step-by-step Jira MCP setup, what it does well, where it burns tokens unnecessarily, and when a simple jq query on a local cache beats the whole thing.
· 6 min read - 03
Setting Up the Notion MCP Server
Notion's MCP server is the fastest path to an agent that can read your docs and write structured pages. The quirks are worth knowing before your first session.
· 6 min read - 04
Automating Release Notes with AI Agents
From git log to structured release notes in Notion and Jira — a real end-to-end automation flow, not a toy example.
· 6 min read - 05
Paperclip: Managing AI Agents Like a Team
One agent helping you code is a power tool. Multiple agents running autonomously is a team. Paperclip is the org chart, the budget, and the audit log for that team.
· 6 min read - 06
OpenClaw: A Personal AI with Eyes and Hands
A smart model with eyes and hands at a desk — that's how one user described it. OpenClaw runs locally, connects to your chat apps, and executes real tasks with real system access.
· 6 min read - 07
Hermes: Self-Improving Agents on Cheap Infrastructure
A self-improving autonomous agent that runs on a $5 VPS, works with any LLM, and follows open standards. Nous Research's Hermes is what model-agnostic agentic infrastructure looks like.
· 7 min read