一键导入
session-memory
Cross-session persistent context — remember project decisions, architecture choices, and team conventions
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
Cross-session persistent context — remember project decisions, architecture choices, and team conventions
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
Detect and clean AI-typical writing patterns in documents — em dashes, filler phrases, over-formality
Route questions to Claude, Codex, or Gemini for multi-perspective answers
Claude-Codex-Gemini tri-model orchestration — decompose, query, synthesize
Health check — verifies omg plugin is correctly installed and working
Quick reference — what omg agents and skills are available and how to use them
About omg — version, architecture, capabilities, and credits
| name | session-memory |
| description | Cross-session persistent context — remember project decisions, architecture choices, and team conventions |
| tags | ["persistence","context","utility"] |
Uses Copilot's native store_memory (confirmed cross-session persistent) plus .omg/ files for structured data.
| Type | store_memory Key | File | Example |
|---|---|---|---|
| Project context | omg:project | .omg/research/project-context.md | "This is a TypeScript CLI tool using tsup + vitest" |
| Architecture decisions | omg:decisions | .omg/research/decisions.md | "We chose TF-IDF over embeddings because..." |
| Team conventions | omg:conventions | .omg/research/conventions.md | "Always use function keyword, not arrow" |
| Active work | omg:active-plan | .omg/plans/*.md | Current plan being executed |
| Active spec | omg:active-spec | .omg/research/spec-*.md | Current spec from deep-interview |
| Last review | omg:last-review | .omg/reviews/*.md | Most recent review verdict |
"Remember: we decided to use PostgreSQL instead of MongoDB because the team has more SQL experience"
→ Appends to .omg/research/decisions.md:
## 2026-04-07: Database Choice
**Decision:** PostgreSQL over MongoDB
**Reason:** Team SQL experience
→ Updates store_memory key omg:decisions with latest entry summary.
"What architecture decisions have we made?"
→ Reads store_memory key omg:decisions for quick summary
→ Reads .omg/research/decisions.md for full history
"What do you remember about this project?"
→ Reads all omg:* keys from store_memory
→ Lists .omg/research/ files
"Forget the database decision"
→ Removes entry from .omg/research/decisions.md
→ Updates store_memory key
Skills automatically save key context:
omg:active-planomg:active-specomg:last-review.omg/qa-logs/store_memory: quick discovery, auto-loaded by Copilot, survives sessions.omg/ files: full data, git-trackable, human-readable, auditableGood: "Remember that src/auth/ uses JWT with 15-minute expiry and refresh tokens stored in httpOnly cookies" → Saves specific, actionable context that future agents can use.
Bad: "Remember everything" → Too vague. What specifically should persist?
.omg/research/ filestore_memory indexed for quick discoverysession memory, save context, what do you remember