← 返回 Skills 市场
hiparker

🧠 EchoFadeMemory

作者 hiparker · GitHub ↗ · v2.0.0 · MIT-0
cross-platform ✓ 安全检测通过
189
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install echo-fade-memory
功能描述
Runs a thin long-term memory workflow on top of the echo-fade-memory service. Use proactively whenever an answer may depend on prior session context, durable...
使用说明 (SKILL.md)

Echo Fade Memory

This skill turns echo-fade-memory into an installed agent memory operating layer.

The public agent contract is intentionally thin:

  • store
  • recall
  • forget

Image memory is folded into the same store/recall/forget contract. Dashboard and debugging routes live under /v1/dashboard/* and are not part of the agent-facing tool surface.

Natural Triggers in OpenClaw

Use this skill implicitly when the conversation includes:

  • remember this / 记住这个
  • what did we decide before / 上次定的是什么
  • user preferences, durable constraints, corrections
  • project decisions worth carrying across sessions
  • screenshots, diagrams, receipts, whiteboards, or UI states that may matter later
  • repeated failures that reveal a reusable workaround
  • elliptical continuity prompts such as 那个、这个、继续刚才的、你知道的
  • time-indexed prompts such as 今天、刚刚、最近、这次、又、还、还是、依然
  • continuity checks such as 你记得吗、你不是知道吗、你忘了?

Prefer over-triggering low-cost recall to under-triggering and answering as if no history exists.

If http://127.0.0.1:8080 is unreachable in a containerized environment, set:

export EFM_BASE_URL=http://host.docker.internal:8080

Quick Reference

Situation Action
Start of a task or session Recall relevant context with ./scripts/recall.sh "\x3Cquery>"
User states a durable preference / decision / correction Store it immediately with ./scripts/store.sh "\x3Ccontent>" --summary "\x3Csummary>" --type \x3Ctype>
User sends an image or screenshot worth keeping Store it with ./scripts/store.sh "\x3Cfile-path>" --object-type image
Need old memory, image, or topic with one query Use ./scripts/recall.sh "\x3Cquery>"
User asks to delete wrong or obsolete memory Use ./scripts/forget.sh "\x3Cquery-or-id>"
Need debug analytics or dashboards Open /dashboard or call /v1/dashboard/*

Core Workflow

1. Recall Before Responding

Before answering about prior decisions, preferences, goals, screenshots, or unresolved issues:

./scripts/recall.sh "database choice for this project"

Inspect:

  • mixed
  • memories
  • images
  • entities

If a recalled memory is fuzzy, you can still ground it with GET /v1/memories/\x3Cid>/ground, but keep that as an internal troubleshooting path rather than the default agent contract.

2. Store Durable Facts Early

When the user says something durable, store it before moving on.

Recommended minimal memory shape:

  • content
  • summary
  • type
./scripts/store.sh \
  "User prefers dark mode and minimal UI" \
  --summary "dark mode preference" \
  --type preference

Advanced fields still exist, but only add them when you have a clear reason:

  • --importance
  • --ref
  • --kind
  • --conflict-group

Use higher importance only for:

  • preferences
  • corrections
  • project decisions
  • constraints
  • explicit "remember this" statements

3. Store Images Through the Same Entry

When the conversation includes a screenshot, whiteboard, receipt, or other durable visual artifact.

Recommended minimal image shape:

  • file_path or url
  • optional caption
  • optional tags
  • optional ocr_text
./scripts/store.sh \
  "/absolute/path/to/meeting-whiteboard.png" \
  --object-type image \
  --caption "meeting whiteboard about rollout" \
  --tag rollout \
  --ocr-text "Deployment Checklist"

Advanced image flags still exist, but they are not the default mental model:

  • --session
  • --kind
  • --actor
  • --memory-id
  • --url

Use image memory when the user is likely to ask:

  • "上次那张图"
  • "有猫那张图"
  • "包含某句话的截图"
  • "和那个决定相关的图片"

4. Forget Wrong or Obsolete State

If a memory or image is incorrect, unsafe, or obsolete:

./scripts/forget.sh "that obsolete deployment note"
./scripts/forget.sh "\x3Cimage-id-or-query>" image

Memory Taxonomy

Situation memory_type Notes
User preference preference Use high importance
Project decision project Add conflict_group for versioning
Goal / pending work goal Good for future follow-ups
Error workaround project Prefix summary with error: or learning:
Capability request goal or project Prefix summary with feature-request:

Scripts

Script Purpose
scripts/health-check.sh Verify the server is reachable
scripts/store.sh Unified store wrapper for memory and image objects
scripts/recall.sh Unified federated recall wrapper
scripts/forget.sh Unified forget wrapper for memory or image objects
scripts/activator.sh Hook reminder for recall/store discipline
scripts/error-detector.sh Hook reminder when command output looks like a failure

Setup

Service Availability

./scripts/health-check.sh

OpenClaw Config

Recommended entry in openclaw.json:

{
  "skills": {
    "entries": {
      "echo-fade-memory": {
        "baseUrl": "http://host.docker.internal:8080"
      }
    }
  }
}

Recommended precedence:

  1. EFM_BASE_URL
  2. skills.entries.echo-fade-memory.baseUrl
  3. default http://127.0.0.1:8080
安全使用建议
This skill appears to do what it says: it posts queries and small JSON payloads to an echo-fade-memory HTTP service and injects a bootstrap reminder into OpenClaw. Before installing, confirm the memory service base URL is trusted (EFM_BASE_URL or local 127.0.0.1). Avoid pointing EFM_BASE_URL at unknown remote servers (that would allow memory contents and queries to be sent off-host). Note that scripts send file path strings for images but do not upload file contents themselves; review your memory service implementation if you want content uploaded. If you don't want automatic prompt injections, don't enable the hook or avoid enabling the activator/error-detector hooks. If you need higher assurance, run the health-check against a local instance, inspect the memory-service server code, or keep the service bound to localhost.
功能分析
Type: OpenClaw Skill Name: echo-fade-memory Version: 2.0.0 The echo-fade-memory skill bundle provides a long-term memory layer for AI agents, enabling them to persist and retrieve user preferences, project decisions, and visual artifacts across sessions. The implementation consists of shell and Python scripts (e.g., store.sh, recall.sh, forget.sh) that interface with a local or remote REST API, alongside OpenClaw hooks (handler.js) that inject usage reminders into the agent's bootstrap process. The code logic is transparent, lacks obfuscation, and aligns strictly with the stated purpose of memory management without exhibiting signs of data exfiltration or unauthorized system access.
能力评估
Purpose & Capability
Name/description describe a thin memory layer and the repository contains only lightweight helpers (store/recall/forget wrappers), a hook that injects a reminder, and docs. The files and the behavior (talking to a base URL for store/recall/forget) align with the stated purpose.
Instruction Scope
Runtime instructions and scripts only perform JSON POST/GET calls to the configured memory service endpoints and emit reminder text. The scripts read two specific OpenClaw config paths (/Users/system/.openclaw/openclaw.json and /home/node/.openclaw/openclaw.json) only to find a configured baseUrl; they do not read arbitrary files or exfiltrate local file contents. Note: store.sh accepts a file path string for image memories but does not itself upload file contents — it sends the path/metadata to the memory service.
Install Mechanism
There is no install spec; the package is instruction-only with small local scripts and a JS hook. Nothing is downloaded or extracted from an external URL, and no package managers are invoked.
Credentials
The skill declares no required environment variables or credentials. It honors an optional EFM_BASE_URL which can point to localhost or another HTTP endpoint. If EFM_BASE_URL is set to an external server, store/recall/forget operations will transmit queries and memory payloads to that endpoint — this is expected for a memory service but users should ensure the endpoint is trusted.
Persistence & Privilege
always:false and no privileged installs are requested. The OpenClaw hook injects a virtual bootstrap reminder (modifies event.context.bootstrapFiles) which is consistent with a hook behavior. The skill can be invoked autonomously by the agent (platform default) — this is normal but raises the usual autonomous-invocation considerations.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install echo-fade-memory
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /echo-fade-memory 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v2.0.0
Major update: Script interface unified and image memory added to core contract. - Added unified script wrappers: `store.sh`, `recall.sh`, and `forget.sh`, replacing prior memory-only scripts. - Image and screenshot memory are now handled alongside text memories using the same workflow and script interface. - Public agent contract is simplified to `store`, `recall`, `forget`; advanced features and dashboards moved out of default workflow. - Natural triggers for memory use in OpenClaw clarified. - Updated documentation for all workflows, including image handling, minimal memory fields, and environment configuration. - Legacy scripts for memory operations removed and replaced by skill-local unified wrappers.
v1.0.0
**Summary:** Version 1.1.0 introduces a comprehensive, workflow-oriented memory layer that replaces multiple reference skills with an integrated, script-driven approach. - New project-native workflow replaces elite-longterm-memory and self-improving-agent scripts with core scripts and hooks. - Clear action table for recall, store, reinforce, ground, and decay operations via included scripts. - Enhanced memory strength and decay model, with practical guidance for setting importance and usage patterns. - Detailed mapping showing how functionality from older skills is replaced by scripts and workflows within this skill. - Expanded documentation: setup instructions, agent guidelines for session management, and troubleshooting tips. - Additional resources and integration guides are now included for easier adoption and extension.
元数据
Slug echo-fade-memory
版本 2.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 2
常见问题

🧠 EchoFadeMemory 是什么?

Runs a thin long-term memory workflow on top of the echo-fade-memory service. Use proactively whenever an answer may depend on prior session context, durable... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 189 次。

如何安装 🧠 EchoFadeMemory?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install echo-fade-memory」即可一键安装,无需额外配置。

🧠 EchoFadeMemory 是免费的吗?

是的,🧠 EchoFadeMemory 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

🧠 EchoFadeMemory 支持哪些平台?

🧠 EchoFadeMemory 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 🧠 EchoFadeMemory?

由 hiparker(@hiparker)开发并维护,当前版本 v2.0.0。

💬 留言讨论