Key Tracker is an intelligent information capture skill that acts as an external brain by automatically extracting deadlines, decisions, and tasks from AI interactions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install key-tracker
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install key-tracker using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Key Tracker: Intelligent Information Capture?
Key Tracker serves as a sophisticated information capture system designed to ensure no critical detail is lost during development or collaboration. Unlike a simple note-taking tool, it functions as an external brain for your AI agent, monitoring conversations for 16 specific types of data including milestones, blockers, and commitments.
By leveraging Openclaw Skills like Key Tracker, developers can maintain a high-fidelity record of project evolution without manual documentation overhead. It provides a traceable decision-making process and an automated reminder system for all outstanding items mentioned during a session.
Key Tracker: Intelligent Information Capture Use Cases
- Tracking project deadlines and milestones mentioned during AI-assisted brainstorming.
- Logging architectural decisions and their rationales for future traceability.
- Automatically capturing commitments made by either the user or the agent during a workflow.
- Identifying and tracking blockers or risks that emerge during a coding session.
- Organizing learning points and sudden insights into a persistent knowledge base.
How Key Tracker: Intelligent Information Capture Works
- The skill monitors incoming conversation streams for specific regex patterns related to 16 key categories like deadlines, decisions, and risks.
- Upon detecting a match (e.g., a date followed by a completion keyword), it parses the context and source text.
- The extracted information is categorized into one of seven primary buckets (Time, Decision, Problem, Commitment, etc.) and assigned a unique ID.
- Data is persisted as structured JSON objects in organized subdirectories within the local environment.
- Users can query the stored records through natural language commands to retrieve specific summaries or status updates.
Key Tracker: Intelligent Information Capture Setup
To get started with this addition to your Openclaw Skills, use one of the following methods:
# Method 1: Install via ClawHub (Recommended)
clawdhub install key-tracker
# Method 2: Manual Installation
git clone https://github.com/vincentlau2046-sudo/key-tracker.git ~/.openclaw/workspace/skills/key-tracker
# Initialize the required record directories
mkdir -p ~/.openclaw/workspace/.keyrecords/{Time,Problem,Decision,Commitment,Process,Knowledge,People}
Key Tracker: Intelligent Information Capture Data Schema & Taxonomy
Key Tracker organizes data into a structured directory tree under ~/.openclaw/workspace/.keyrecords/. Each entry follows a consistent JSON format for interoperability with other Openclaw Skills.
| Category | File Examples | Purpose |
|---|---|---|
| Time | deadlines.json, milestones.json |
Schedules and delivery dates |
| Problem | issues.json, blockers.json, risks.json |
Obstacles and warnings |
| Decision | decisions.json, rationales.json |
Project choices and justifications |
| Knowledge | learnings.json, insights.json |
PKM and technical notes |
Record Object Schema:
id: Unique identifier (e.g., KR-20260312-001)type: The specific sub-type (e.g., deadline)context: Summary of the surrounding discussionsource_text: The raw snippet that triggered the entrystatus: Current state (pending/resolved)priority: Importance level (high/medium/low)
Key Tracker: Intelligent Information Capture Advanced Features
- Automated Regex detection for sophisticated pattern matching across 16 different data types.
- Scheduled review tasks: Configure Morning Reminders (08:00) for deadlines and Evening Summaries (21:00) for daily logs.
- Multi-category support: Capture complex entries that link a decision directly to its rationale or an identified risk.
- Natural Language Querying: Ask the agent 'What decisions have we made?' or 'What is blocking us?' to get instant structured reports.
- Persistent JSON storage: Enables easy backup, version control via Git, and data portability between different Openclaw Skills instances.
SKILL.md
Loading
Related Openclaw Skills
SkillForge
SkillForge is a specialized generator that builds high-quality, production-grade AI agent skills through a structured 7-step optimization pipeline.
Sales Report Parser
A high-performance toolkit for extracting structured sales data from report images using OCR and AI-driven JSON parsing.
Planner
A local-first planning engine designed to transform vague objectives into structured roadmaps, itineraries, and actionable next steps.
PPT Audio To Video
A specialized tool for converting presentation slides and narration audio into high-quality narrated video files using AI transcription and automated rendering.
Prospector
Prospector is an automated lead generation engine that identifies potential client companies based on keywords and extracts verified contact emails from web sources.
Cron
A local-first recurring schedule engine for Openclaw Skills that converts natural language timing into structured execution plans.
Featured*
