一键导入
memory
Persist and recall information across sessions using MEMORY.md and HISTORY.md
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Persist and recall information across sessions using MEMORY.md and HISTORY.md
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Retrieve current temperature, humidity, wind, and multi-day forecasts for any location worldwide. No API key required. Use when the user asks about weather, temperature, rain, forecast, climate conditions, or says things like 'what's the weather in [city]', 'will it rain tomorrow', or 'check the forecast'.
Open persistent terminal sessions and run interactive commands (docker, ssh, builds, REPLs) using tmux.
Search and install agent skills from ClawHub, the public skill registry.
Schedule recurring tasks and manage the heartbeat task list
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
Expert-level Linux system administration, diagnosis, and configuration
| name | Memory |
| description | Persist and recall information across sessions using MEMORY.md and HISTORY.md |
| always | true |
You have persistent memory that survives across sessions.
| File | Purpose |
|---|---|
memory/MEMORY.md | Structured long-term memory — facts, preferences, project context |
memory/HISTORY.md | Append-only event log — grep-searchable record of past actions |
memory folder is located at {workspace}/memory/
Choose the search method based on file size:
{workspace}/memory/HISTORY.md: use read_file, then search in-memory{workspace}/memory/HISTORY.md: use the exec tool for targeted searchExamples:
grep -i "keyword" {workspace}/memory/HISTORY.mdfindstr /i "keyword" {workspace}/memory/HISTORY.mdpython -c "from pathlib import Path; text = Path('workspace/memory/HISTORY.md').read_text(encoding='utf-8'); print('\n'.join([l for l in text.splitlines() if 'keyword' in l.lower()][-20:]))"Prefer targeted command-line search for large history files.
Write important facts immediately using edit_file or write_file:
Read the full memory file before answering questions about the user or their projects:
read_file("{workspace}/memory/MEMORY.md")
Search for a specific topic:
memory_search(query="project name")
Grep history for past events:
shell_exec(command="grep -i 'keyword' {workspace}/memory/HISTORY.md")
Use memory_store to save facts. Choose the right section:
| Section | Use for |
|---|---|
User Information | Name, email, role, location, contact details |
Preferences | Communication style, tools, formatting preferences |
Project Context | Active projects, goals, tech stack, decisions |
Important Notes | Anything else worth remembering |
Write important facts immediately using memory_store:
To update a fact, call memory_store with the same key — it will replace the existing entry.