A meta-skill that ensures your AI coding agent operates with complete situational awareness by automatically loading relevant project context files.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install auto-context
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 auto-context 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 Auto-Context?
Auto-Context is a foundational meta-skill designed to prevent the common pitfall of AI agents operating in a vacuum. By implementing a strict situational awareness protocol, this skill mandates that the agent reads and synthesizes critical project documentation—such as TODO lists, roadmaps, and handoff notes—before executing any significant command. This approach significantly reduces wasted effort, prevents duplicate work, and ensures that every implementation aligns perfectly with the current project state and long-term goals.
By incorporating these Openclaw Skills into your development environment, you ensure that your AI assistant is always up to date with recent changes and upcoming priorities. It acts as an automated 'pre-flight check' for the AI, building a reliable mental model of the project's history and future direction without manual intervention from the developer.
Auto-Context Use Cases
- Resuming a coding session after a break to rebuild the AI's mental model.
- Starting a new feature implementation to ensure constraints and priorities are understood.
- Debugging complex issues by reviewing recent findings and changelog entries.
- Performing a code refactor with full awareness of planned architectural changes.
- Preparing for a project handoff by ensuring all documentation is synthesized for the next phase.
How Auto-Context Works
- The protocol triggers automatically when the AI detects a major action like starting a task, refactoring, or planning.
- It prioritizes loading critical files such as TODO.md and roadmap.md to establish baseline project goals.
- High-priority and task-specific files, such as implementation plans or recent handoffs, are fetched based on existence.
- The system performs a staleness check on file modification dates using system-level commands to warn about outdated info.
- A concise context summary is presented to the user, providing a synthesized view of the project phase, active milestones, and blockers.
- The primary AI command then executes with full awareness of the loaded context.
Auto-Context Setup
To integrate this situational awareness protocol into your workflow, install it via the CLI:
npx clawhub@latest install auto-context
Ensure your project contains standard documentation files like TODO.md or roadmap.md in the root or directories like /docs or /.cursor to allow the skill to function at its peak.
Auto-Context Data Schema & Taxonomy
Auto-Context scans and organizes project metadata according to a strict priority taxonomy:
| Priority | File Types | Metadata Extracted |
|---|---|---|
| Critical | TODO.md, roadmap.md | Current task backlog, phase status, and milestones. |
| High | task_plan.md, handoffs/*.md | Active implementation steps and recent transition notes. |
| Medium | findings.md, CHANGELOG.md | Research discoveries, logic decisions, and recent changes. |
| Low | sessions/*.md | Historical session summaries for deep state recovery. |
Auto-Context Advanced Features
- Automated staleness detection providing warnings for files updated more than 7 days ago.
- Intelligent fallback path detection searching for context files in /docs, /ai, and /.cursor directories.
- Proactive context synthesis that produces a tight executive summary instead of raw file dumps.
- Deep integration with other Openclaw Skills to provide the 'understand' phase of any complex workflow.
- Execution order enforcement that ensures context is always loaded before primary skills proceed.
SKILL.md
Loading
Related Openclaw Skills
Auth Patterns
A comprehensive guide to implementing industry-standard authentication and authorization patterns including JWT, OAuth 2.0, and session management.
Assess Risk for Uniswap
A powerful risk evaluation tool for Uniswap that provides independent security audits for swaps, liquidity positions, and bridge operations.
Article Illustrator
A specialized agent skill that analyzes written content to generate and insert consistent, contextually relevant illustrations using a unique Type x Style framework.
Architecture Patterns
A comprehensive guide to implementing Clean Architecture, Hexagonal, and DDD patterns for robust backend systems using Openclaw Skills.
Event Store
A comprehensive technical framework for architecting, implementing, and managing immutable event stores in event-sourced applications.
Batch Swap
A high-performance trading skill for executing sequential token swaps and automated portfolio rebalancing with independent safety checks.
Featured*
