---
title: "Discussion on \"How SoulClaw Cuts Token Usage by 60% With Tiered Bootstrap Loading\" | Hashnode"
description: "Discussion on \"How SoulClaw Cuts Token Usage by 60% With Tiered Bootstrap Loading\". Every time your AI agent responds, the runtime stuffs its entire workspace into the system prompt. SOUL.md. IDENTITY.md. AGENTS.md. TOOLS.md. USER.md. MEMORY.md. Every file in memory/. HEARTBEAT.md. BOOTSTRAP.md.\nThat's thousands of tokens per turn, ..."
url: "https://hashnode.com/posts/how-soulclaw-cuts-token-usage-by-60-with-tiered-bootstrap-loading/69abd502709686a2872b3e27"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [How SoulClaw Cuts Token Usage by 60% With Tiered Bootstrap Loading](https://clawsouls.hashnode.dev/how-soulclaw-cuts-token-usage-by-60-with-tiered-bootstrap-loading?utm_source=hashnode&utm_medium=feed)

Every time your AI agent responds, the runtime stuffs its entire workspace into the system prompt. SOUL.md. IDENTITY.md. AGENTS.md. TOOLS.md. USER.md. MEMORY.md. Every file in memory/. HEARTBEAT.md. BOOTSTRAP.md. That's thousands of tokens per turn, ...

Mar 7 · 4 min read · clawsouls.hashnode.dev  
Read the article

-   [#context-window](https://hashnode.com/tag/context-window)
-   [#openclaw](https://hashnode.com/tag/openclaw)
-   [#progressive-disclosure](https://hashnode.com/tag/progressive-disclosure)
-   [#soul-spec](https://hashnode.com/tag/soul-spec)
-   [#soulclaw](https://hashnode.com/tag/soulclaw)
-   [#system-prompt](https://hashnode.com/tag/system-prompt)
-   [#token-optimization](https://hashnode.com/tag/token-optimization)

## Written by

[ClawSouls](https://hashnode.com/@clawsouls)

@clawsouls · Seoul, South Korea

  
[View profile](https://hashnode.com/@clawsouls)

Open persona spec for AI agents — clawsouls.ai (http://clawsouls.ai/)

## Responses

Markdown supported · ⌘↵ to post

No responses yet.