Skills Plugins MCP Prompt Model 博客 我的中心

opencode-memory

Persistent memory system with .memory.md, compaction, and cross-project sync

DeepseekModel 官方收录技能 质量 良好 · 48 v1.0.0

获取

https://deepseekmodel.com/api/download.php?id=digi4care-opencode-mastery-src-skill-opencode-memory-skill-md&format=skill
下载 .skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用
.skill 文件中 system_prompt 字段的实际内容。
name opencode-memory description Persistent memory system with .memory.md, compaction, and cross-project sync license MIT compatibility opencode metadata {"author":"OpenCode Community","version":"1.0.0"} triggers ["memory",".memory.md","onthouden","remember","persistent memory","context onthouden","wat weet je","remember this","tell me about"] OpenCode Memory System Persistent memory keeps useful project context available across sessions via the memory plugin, tools, hooks, and slash commands. Quick Commands /memory status Show memory health and indexing status /memory sync [--force] [--verbose] Index MEMORY.md and related sources /memory compact Reduce memory size while preserving key context /remember "text" [flags] Store new memory with category/priority metadata /recall "query" [flags] Search memory with citations and relevance scores /forget "query" [--force] Remove a memory entry (confirmation by default) See full command behavior and examples in docs/MEMORY_USAGE.md . Usage Patterns Starting a New Project Run /memory status to confirm memory is enabled and healthy. Add baseline context with /remember "Project uses Bun + strict TypeScript" --category context . Continue saving decisions as they happen (architecture, standards, gotchas). Finding Past Decisions Use targeted recall queries: /recall "decision about auth provider" /recall "why vectorWeight 0.7" --source memory --min-score 0.3 The expected result format is citation + score + preview, for example: MEMORY.md#L45-L52 (0.87): "Use verb routing for /memory commands to keep UX consistent..." Cleaning Up and Maintenance Run /memory sync after substantial MEMORY.md edits. Run /memory compact when memory grows noisy or repetitive. Use /memory sync --force if search quality drops after major edits. Best Practices Writing Effective Memories Be specific about what changed and why it matters. Include enough context to be reusable later. Prefer one clear memory per idea. Good: Use memory_search as shared entry point for /recall and /forget to keep filters consistent. Too vague: Search was updated. Use Categories Consistently decision for architecture choices and tradeoffs. pattern for repeatable implementation approaches. correction for fixes and lessons learned. preference for team/user style constraints. context for project facts and constraints. When to Use Priority high : security, production stability, irreversible decisions. medium : most implementation notes. low : temporary reminders and lightweight context. Organization Tips Use consistent wording for recurring topics to improve recall matches. Periodically compact to keep retrieval relevant. Prefer concise technical language over conversational phrasing. Integration Tips Memory commands are thin wrappers around plugin tools ( memory_status , memory_sync , memory_search , memory_remember ). Hooks automatically bootstrap and persist useful context at session boundaries. Configuration is shared through opencode.config.yaml and src/lib/config/ , so memory behavior stays aligned with project-wide settings. Use this skill as quick guidance; use docs/MEMORY_USAGE.md for complete installation, configuration, troubleshooting, and workflow coverage. Difficulty : ⭐⭐ (Intermediate) Complexity : Medium (scripts + integration) Use Case : Persistent context across sessions, project memory, cross-project knowledge
Agent 识别该技能的关键词,点击任意一个即可复制。

该技能未提供触发词。

下载的 .skill 包内含以下字段。
字段 说明
format格式标识(skill/v1)
skill_id技能唯一 ID
name技能名称
version版本号
description技能描述
category所属分类(数组)
trigger_words触发词列表
tags标签列表
source来源标识
source_url来源链接(本页地址)
exported_at导出时间(每次下载生成)
system_prompt系统提示词正文
model_config模型参数:provider / model / temperature / max_tokens / top_p
examples示例
install_guide各平台导入说明(Coze / Dify / Claude / 自定义框架)
同一份技能可按不同平台格式导出。
.skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用 下载
.skillpro 增强格式,额外含脚本 / 工具 / 依赖 / 钩子占位 下载
.json 纯 JSON 导出,只含 system_prompt 与模型参数 下载
Coze 带 frontmatter 的 Markdown,Coze 平台导入用 下载
Dify Dify DSL,创建应用后直接导入 下载

每日精选 Skill 推荐,免费送到你邮箱

输入邮箱,每天接收一个精选 AI Agent 技能推荐。完全免费,持续更新。

验证码 --

提交后我们会发送一封确认邮件,点击邮件里的链接才会开始收信。

完全免费,取消任意时间。我们不会发送垃圾邮件。