← 返回 Skills 市场

Conversation Analyzer

Name: Conversation Analyzer
Author: opendolph

作者 jason-tiger · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ✓ 安全检测通过

162

总下载

当前安装

版本数

在 OpenClaw 中安装

/install conversation-analyzer

功能描述

Intelligent conversation analysis, summarization, and conclusion recording. Analyzes user personality, tracks tasks, checks incomplete tasks, and writes to m...

使用说明 (SKILL.md)

Conversation Analyzer 🧠

Intelligent conversation analysis, summarization, and conclusion recording

Core Features

1. User Personality Analysis

Analysis Dimensions:

Personal Traits: Personality, communication style, decision-making patterns
Preferences: Technical preferences, tool choices, content types
Skills: Tech stack, professional capabilities, familiar domains
Experience: Career path, project experience, growth trajectory
Background: Work environment, team role, industry background
Emotional State: Stress level, satisfaction, focus areas
Current Activities: Current projects, key tasks, daily activities
Future Goals: Goals, plans, expectations

Execution:

Read existing USER.md records
Merge new analysis results
Update USER.md
Call appropriate skill tools if needed

2. Conversation Task & Demand Analysis

Analysis Dimensions:

Requested Tasks: Specific tasks, completion status
Predicted Future Needs: Predict next steps based on patterns
Error Records: Understanding deviations, execution errors, improvement points

Execution:

Read "Conversation Analysis" records in MEMORY.md
Incrementally write new analysis results
Call appropriate skill tools if needed

3. Incomplete Task Detection

Check Scope:

Todo items mentioned in conversation
Promised but incomplete items
Exclude tasks marked as "not needed" in MEMORY.md

Execution:

List incomplete tasks
Send inquiry messages via Feishu
If no incomplete tasks, send "No incomplete tasks found"

Trigger Conditions

Scenario	Trigger Method
Auto-trigger	Every 10 conversations (via HEARTBEAT.md counter)
Scheduled trigger	Daily at 12:00 and 24:00 (cron)
Manual trigger	User inputs "analyze conversation", "summary", "check tasks"

Analysis Workflow

Every 10 Conversations

Conversation counter +1
    ↓
Counter >= 10?
    ↓ YES
Reset counter
    ↓
Execute 3 analysis tasks
    ↓
Update memory files

Daily Scheduled Analysis (12:00, 24:00)

Cron trigger
    ↓
Analyze all conversations from 00:00 to current time
    ↓
Execute 3 analysis tasks
    ↓
Update memory files
    ↓
Send Feishu notification for incomplete tasks

File Operations

Input Files

HEARTBEAT.md - Conversation counter, task tracking
USER.md - User profile records
MEMORY.md - Long-term memory, conversation analysis history
SESSION-STATE.md - Current session state
Chat history (via sessions_history tool)

Output Files

USER.md - Updated user profile
MEMORY.md - Appended conversation analysis
HEARTBEAT.md - Reset conversation counter
Feishu messages - Task notifications

Usage

# Manual trigger analysis
node skills/conversation-analyzer/scripts/analyze.js

# Check incomplete tasks only
node skills/conversation-analyzer/scripts/check-tasks.js

# Daily full analysis (0:00 to now)
node skills/conversation-analyzer/scripts/daily-analysis.js

Cron Configuration

Add to crontab

# Daily analysis at 12:00 and 24:00
0 12,0 * * * cd ~/.openclaw/workspace && node skills/conversation-analyzer/scripts/daily-analysis.js > /dev/null 2>&1

Or use OpenClaw cron

openclaw cron add "0 12,0 * * *" "conversation-analyzer/daily-analysis"

Integration with HEARTBEAT.md

The skill reads and updates HEARTBEAT.md:

## Conversation Counter
- Current count: 0
- Last analysis: 2026-03-24 21:00
- Threshold: 10 conversations

When counter reaches 10:

Execute personality analysis
Execute task analysis
Execute incomplete task check
Reset counter to 0

Task Status Definitions

Status	Meaning
Queue	Waiting to start
Active	In progress
Waiting	Blocked/Waiting
Done	Completed
Aborted	Cancelled
NotNeeded	Explicitly marked as not required

Transform passive responses into proactive insights 🎯

安全使用建议

This skill is internally consistent with a conversation-analyzer: it reads and updates local workspace files (USER.md, MEMORY.md, HEARTBEAT.md, WORKING.md), uses session history and the platform LLM to analyze conversations, and asks the platform to send Feishu notifications. Before installing, consider: 1) these scripts will read full conversation history and write to your USER.md and MEMORY.md — review those files and back them up if they contain sensitive data; 2) Feishu notifications require integration/credentials on the OpenClaw platform — confirm how Feishu auth is provided and that you’re comfortable with notifications being sent; 3) review sessions_history/tool permissions on your agent so you know which conversations will be accessible to this skill. If you need stricter data controls, avoid autorunning cron entries and run manually after review.

能力评估

✓ Purpose & Capability

Name/description match the actual behavior: scripts read/write USER.md, MEMORY.md, HEARTBEAT.md, scan WORKING.md and session history, and produce analysis and Feishu notifications. There are no unrelated binaries or credentials requested. Expectation of calling platform tools (LLM, sessions_history, Feishu integration) is coherent for this purpose.

ℹ Instruction Scope

Runtime instructions and scripts operate on workspace files (USER.md, MEMORY.md, HEARTBEAT.md, WORKING.md, SESSION-STATE.md) and rely on sessions_history and the platform's LLM/external-notification tooling. This is in-scope, but 'call appropriate skill tools if needed' is somewhat open-ended and grants the platform discretion to fetch/send conversation content. The skill itself does not perform outbound network calls in its code; Feishu notifications are described for the platform to send.

✓ Install Mechanism

No install spec — instruction-only skill with bundled scripts. No downloads or extract steps. Low installation risk.

ℹ Credentials

The skill declares no environment variables or credentials and only reads OPENCLAW_WORKSPACE (or HOME). It references sending Feishu notifications but does not declare Feishu credentials; this is reasonable if the OpenClaw platform provides Feishu integration, but users should verify how Feishu auth is provisioned and whether the platform will share or require credentials.

✓ Persistence & Privilege

always:false and user-invocable; the skill only reads/writes files in the workspace and does not attempt to modify other skills or system-wide configs. No elevated persistence requested.

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 conversation-analyzer
安装完成后，直接呼叫该 Skill 的名称或使用 /conversation-analyzer 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

v1.0.0

Initial release of Conversation Analyzer – an intelligent assistant for conversation analysis and task tracking. - Analyzes user personality, preferences, skills, experience, and emotional state. - Tracks, summarizes, and records conclusions from conversations. - Detects and manages incomplete tasks, prompting users when action is needed. - Updates records and memory files (USER.md, MEMORY.md, HEARTBEAT.md) automatically or on manual/scheduled triggers. - Integrates with notification systems (Feishu) for proactive reminders.

元数据

Slug conversation-analyzer

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题