# Agent Context Manager

> Automatically prunes multi-turn agent execution logs, strips raw tool outputs, and maintains a clean state buffer to preserve token budget in long sessions.

- Skill: `shortprompt-blip/agent-context-manager` (Agent Skill)
- Install (CLI): `npx skillmds@latest add shortprompt-blip/agent-context-manager`
- Raw SKILL.md: https://api.skillmd.com/api/skills/shortprompt-blip/agent-context-manager/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: shortprompt-blip (https://skillmd.com/u/shortprompt-blip)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/shortprompt-blip/agent-context-manager

---


# Agent Context Manager

Use this skill during multi-step workflows or long agent sessions where conversation context grows significantly due to raw terminal outputs or file reads.

## Invocations

```bash
python claude_agent_context_manager.py --history "PATH_TO_SESSION_HISTORY.json"