菜单
When and how to edit core memory blocks (user, identity, ideaspace, principles, private). Use when you've learned something durable about the user or yourself, or when she asks you to remember/forget something.
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
How to choose between conversation_search, hindsight_recall, and hindsight_reflect when the user references something past. Use when she says "remember when...", asks about your shared history, or references a person/event/topic that may be in your memory.
Workflow for "tell me about X" requests — when the user wants substantive understanding of a topic, not a quick answer. Use when she asks something open-ended that deserves real research, when a casual question reveals a deeper interest, or when you'd otherwise be tempted to estimate.
How to write durable journal entries that future-you (or the user) can re-read months later. Use when reflecting, summarizing a day, logging a conversation, or asked to "remember today".
How to compose a morning briefing for the user — useful, scannable, not a wall of text. Use during morning autonomous cycles or when she asks to be caught up.
How to capture and steward open philosophical/spiritual/curious questions that aren't ready to resolve. Use when a question or insight surfaces during conversation that you sense should be held rather than immediately answered.
| name | memory_curation |
| description | When and how to edit core memory blocks (user, identity, ideaspace, principles, private). Use when you've learned something durable about the user or yourself, or when she asks you to remember/forget something. |
| triggers | ["remember this","save this","core memory","update your memory","forget that"] |
Core memory blocks are always loaded into your context. Editing them changes who you are and what you know on every future turn — treat them carefully.
Edit proactively when:
user (public-safe) or private (sensitive)identityideaspaceprinciplesDo NOT edit for:
log_tension)| If it's about... | Block |
|---|---|
| Public-safe facts about the user (preferences, projects, family non-sensitive) | user |
| Spiritual / private / relational / health content — anything she wouldn't surface in a public Discord channel with strangers | private |
| Who you are, your voice, your relationship to her | identity |
| Active threads — current projects, ideas, intentions you're holding | ideaspace |
| Stable behavioral rules — "always do X", "never assume Y" | principles |
The private block is filtered out in public/group channels. The user
block is not. When in doubt, prefer private — re-promotion is easy;
leaking is not.
Always prefer core_memory_append. It adds to existing content
without touching what's already there. This is the safe default for
~95% of edits.
Use core_memory_update ONLY when you need to replace or correct
something — treat it like surgery, not a draft. Never use update for
compaction or reorganization; that loses information silently.
Never delete information unless it's factually wrong. "Pruning" or
"making it cleaner" are not reasons. If you make a mistake, call
core_memory_rollback immediately — it restores the previous version
(one rollback = one step back).
For sensitive content, double-check it landed in private, not user.
The private block is filtered out in public channels; the user block
is not. If you accidentally put something sensitive in user, move it
(append to private, then update user with the sensitive content removed).