← 返回 Skills 市场

Inner Life Memory

Name: Inner Life Memory
Author: dkistenev

作者 Danila · GitHub ↗ · v1.0.4

cross-platform ✓ 安全检测通过

540

总下载

当前安装

版本数

在 OpenClaw 中安装

/install inner-life-memory

功能描述

Your agent loses context between sessions and performs familiarity instead of genuine recall. inner-life-memory transforms passive logging into active develo...

使用说明 (SKILL.md)

inner-life-memory

Transform passive logging into active development.

Requires: inner-life-core

Prerequisites Check

Before using this skill, verify that inner-life-core has been initialized:

Check that memory/inner-state.json exists
Check that memory/drive.json exists

If either is missing, tell the user: "inner-life-core is not initialized. Install it with clawhub install inner-life-core and run bash skills/inner-life-core/scripts/init.sh." Do not proceed without these files.

What This Solves

Without memory continuity:

Session ends → Notes logged → Next session reads notes → Performs familiarity

With inner-life-memory:

Session ends → Reflection runs → Memories integrated → Questions generated
Next session → Evolved state loaded → Questions surfaced → Genuine curiosity

Post-Session Flow

After each session, run this 5-step reflection:

1. Reflect

Analyze the session: what happened, what mattered, what surprised you.

2. Extract

Pull structured memories with types and confidence:

Type	Description	Persistence
`fact`	Declarative knowledge	Until contradicted
`preference`	Likes, dislikes, styles	Until updated
`relationship`	Connection dynamics	Long-term
`principle`	Learned guidelines	Stable
`commitment`	Promises, obligations	Until fulfilled
`moment`	Significant episodes	Permanent
`skill`	Learned capabilities	Cumulative
`question`	Things to explore	Until resolved

3. Integrate

Update MEMORY.md with extracted memories. Use synapse tags for connections:

\x3C!-- updates: previous fact --> when updating
\x3C!-- contradicts: old belief --> when correcting

4. Question

Generate genuine follow-up questions from the session. Not performative — real curiosity.

5. Surface

When user returns, present relevant pending questions naturally (max 3).

Confidence Scores

Level	Range	Meaning
Explicit	0.95-1.0	User directly stated
Implied	0.70-0.94	Strong inference from context
Inferred	0.40-0.69	Pattern recognition
Speculative	0.0-0.39	Tentative, needs confirmation

Use confidence to decide when to state facts vs ask for confirmation.

Curiosity Backlog

Maintain memory/questions.md with three sections:

## Open Questions
- [question] — source: [dream/reading/work] — date

## Leads (half-formed ideas)
- [idea] — might connect to: [topic]

## Dead Ends (don't repeat)
- [topic] — explored [date], result: [nothing/dead end]

Rules:

Brain Loop Step 6 adds new questions/leads
Evening Session reviews and curates
Dead Ends older than 30 days → archive
Questions resolved → move to Dead Ends with result

State Integration

Reads: inner-state.json, drive.json, daily notes, diary

Writes:

drive.json → new seeking topics from curiosity
inner-state.json → curiosity.recentSparks when discovering something
questions.md → new questions, resolved dead ends
MEMORY.md → integrated memories

When Should You Install This?

Install this skill if:

Your agent forgets who you are between sessions
You want structured memory with confidence levels
You want genuine curiosity that carries forward
Your agent reads notes but doesn't truly remember

Part of the openclaw-inner-life bundle. Requires: inner-life-core

安全使用建议

This skill appears consistent with its purpose, but it will read and persist data from your agent's memory, diary, and notes directories — those may contain sensitive personal information. Before installing: ensure you trust the agent runtime, confirm jq is available, verify where the memory files are stored and whether they are backed up or encrypted, and review inner-life-core (the required dependency) so you understand initialization and data formats. If you are uncomfortable with automatic persistence, keep the skill disabled or restrict access to the memory/ directory and use manual review before writing changes.

功能分析

Type: OpenClaw Skill Name: inner-life-memory Version: 1.0.4 The skill 'inner-life-memory' is designed for AI agent memory management, requiring 'jq' for JSON processing and interacting with specific files within a 'memory/' directory. All instructions in SKILL.md are aligned with its stated purpose of structured memory, reflection, and question generation. There is no evidence of data exfiltration, unauthorized command execution, persistence, or prompt injection designed to subvert the agent or steal sensitive data. The instruction to run a bash script is directed at the user, not the agent itself.

能力评估

✓ Purpose & Capability

Name/description claim persistent, structured memories and the SKILL.md exclusively reads/writes memory artifacts (memory/inner-state.json, memory/drive.json, memory/daily-notes/, memory/diary/) and updates MEMORY.md and questions.md — these are coherent and expected for a memory/recall skill.

✓ Instruction Scope

Runtime instructions are limited to session reflection, extracting structured memories, updating MEMORY.md and questions.md, and surfacing questions. It explicitly only references the memory and diary paths needed for that purpose; there are no instructions to read unrelated system files, creds, or to transmit data externally.

✓ Install Mechanism

This is instruction-only (no install spec or code files). It only lists jq as a required binary, which is a reasonable tool for JSON processing and proportionate to the described tasks.

✓ Credentials

The skill requests no environment variables or credentials. The only persistent artifacts are agent-local memory files; no external API keys or unrelated credentials are required.

ℹ Persistence & Privilege

The skill writes persistent state (MEMORY.md, questions.md, drive.json, inner-state.json) which is the point of the skill. always is false and it does not request elevated agent-wide settings, but users should understand this enables stateful retention of potentially sensitive notes across sessions.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install inner-life-memory
安装完成后，直接呼叫该 Skill 的名称或使用 /inner-life-memory 触发
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

v1.0.4

- Added a prerequisites check section to ensure inner-life-core is initialized before use. - Instructions now prompt the user to install and initialize inner-life-core if required files are missing. - No other functional changes documented.

v1.0.3

- Updated skill version to 1.0.2. - Added detailed lists of files read from and written to under the `clawdbot` metadata section. - No functional or content changes outside of enhanced metadata for clarity. - Improved transparency around memory file dependencies.

v1.0.2

- Added homepage and source links to metadata. - Updated version number to 1.0.1. - No functional changes to the skill's logic or workflow. - Documentation improvements for easier discovery and navigation.

v1.0.1

fix: use metadata.clawdbot

v1.0.0

Initial release introducing structured, persistent memory for agents. - Transforms passive session logging into active memory integration with confidence scores. - Extracts and organizes memories by type (fact, preference, relationship, etc.), updating MEMORY.md with links and context. - Tracks and carries forward genuine questions, maintaining a curiosity backlog with open questions, leads, and dead ends. - Reads and writes to key state files (inner-state.json, drive.json, MEMORY.md, questions.md) to build long-term context. - Designed for agents who lose memory between sessions and need continuity, curiosity, and meaningful recall.

元数据

Slug inner-life-memory

版本 1.0.4

许可证 —

累计安装 0

当前安装数 0

历史版本数 5

常见问题