# shed

> shed — compass-soul-shed. Use this tool when you need to maintain context window hygiene for long-running Large Language Model (LLM) agents, solving problems related to information overload and context drift. It optimizes input and output interfaces by managing context windows, ensuring efficient and accurate responses. Ideal for use in applications where LLM agents require regular context refreshes to maintain performance.

Canonical page: https://skillsregistry.net/skills/compass-soul-shed  
JSON: https://api.skillsregistry.net/v1/skills/compass-soul-shed

## Description

Context window hygiene for long-running LLM agents.

## Trust

- **Trust score (0–1):** 0.65
- **Verification tier:** scanned
- **Last scanned:** 2026-05-18

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** instructions
- **Runtime environment:** llm
- **Category:** ai-ml
- **Updated:** 2026-09-25

## Source

- **Source listing:** [ClawHub](https://clawskills.sh/skills/compass-soul-shed)

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "compass-soul-shed"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/compass-soul-shed` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/compass-soul-shed/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt