Themata.AI
Themata.AI

Popular tags:

#developer-tools#ai-agents#llms#claude#code-generation#ai-ethics#openai#ai-safety#anthropic#open-source

AI is changing the world. Don't stay behind. Clear summaries, community insight, delivered without the noise. Subscribe to never miss a beat.

© 2026 Themata.AI • All Rights Reserved

Privacy

|

Cookies

|

Contact
ai-agentsdeveloper-toolscost-optimizationcli-tools

Making MCP cheaper via CLI

I Made MCP 94% Cheaper (And It Only Took One Command)

kanyilmaz.me

February 25, 2026

3 min read

Summary

MCP agents incur high costs due to the inclusion of an extensive tool catalog in JSON Schema format. A CLI generated from an MCP server using CLIHub performs the same functions at a significantly reduced cost.

Key Takeaways

  • CLI reduces token usage by approximately 94% compared to MCP by using a lightweight skill listing instead of pre-loading full tool schemas.
  • The session start cost for MCP is around 15,540 tokens, while CLI only requires about 300 tokens.
  • Anthropic's Tool Search, while more efficient than MCP, still incurs higher costs than CLI, making CLI the cheaper option for tool discovery and execution.
  • The author created CLIHub, a directory of CLIs for agent use, and open-sourced a converter to generate CLIs from MCPs with a single command.

Community Sentiment

Mixed

Positives

  • The CLI approach offers practical benefits for token reduction, which can enhance efficiency in AI interactions by minimizing unnecessary context bloat.
  • Existing paradigms like Skills effectively mitigate context bloat and selection overhead, showcasing a more structured approach to tool discovery in AI applications.

Concerns

  • MCP tools contribute significantly to context bloat, as they require sending extensive tool catalogs with every request, which can hinder performance.
  • The claim that MCP tools need to dump the entire tool catalog into the conversation is outdated, as more advanced clients like Claude Code have improved this process.
Read original article

Source

kanyilmaz.me

Published

February 25, 2026

Reading Time

3 minutes

Relevance Score

63/100

🔥🔥🔥🔥🔥

Why It Matters

This page is optimized for focused reading: quick context up top, a clean summary block, and a direct path to the original source when you want the full story.