Context Brief for Openclaw

Context Brief is a specialized memory management tool that enables Openclaw Skills to persist critical session context and decisions across long-running conversations.

tommot2
v3.0.0
Apr 3, 2026
1
1.1k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install context-brief

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install context-brief 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 Context Brief?

Context Brief provides a robust framework for persistent context survival within the AI agent ecosystem. By writing file-based anchors to a dedicated memory directory, it ensures that critical decisions, pending tasks, and project states are preserved even during session resets or context window compaction. This tool acts as a manual and semi-automated checkpointing system that allows Openclaw Skills to maintain a high level of continuity.

The skill is designed with a read-heavy approach to global memory files like MEMORY.md, but it strictly isolates its write operations to a specific anchors folder. This ensures that the agent's core memory remains clean while providing a searchable, timestamped history of development progress and logic flows.

Context Brief Use Cases

  • Preventing context loss during long development sessions or large-scale refactors.
  • Resuming complex tasks after a period of inactivity or system reboot.
  • Creating manual checkpoints before major architectural changes using save context commands.
  • Restoring the specific state of a task when the agent begins to lose track of earlier decisions.

How Context Brief Works

  1. The skill monitors for user triggers such as save context, context check, or inquiries about what the agent remembers.
  2. Upon activation, the agent reads current state data from MEMORY.md, daily logs, and active conversation history.
  3. The skill categorizes information into three levels: brief notes, standard anchors, or deep anchors based on complexity.
  4. A structured Markdown file is generated and saved strictly within the memory/anchors/ directory with ISO 8601 timestamps.
  5. During future sessions or after context compaction, the skill automatically checks for recent anchors to restore missing project state and inform the user.

Context Brief Setup

To install the skill and integrate it into your environment, run the following command via the Openclaw Skills CLI:

clawhub install context-brief

You can also install the full productivity suite for enhanced capabilities:

clawhub install context-brief cross-check setup-doctor locale-dates

Context Brief Data Schema & Taxonomy

Context Brief maintains a strict file hierarchy and data format to ensure interoperability between Openclaw Skills.

Feature Specification
Storage Location memory/anchors/YYYY-MM-DDTHH-mm.md
Format Markdown (Max 20 lines)
Security Automatic redaction of API keys, passwords, and secrets
Metadata Includes ISO 8601 timestamps and task status tags

Each anchor file includes sections for Active State, Decisions, Pending Actions, and Key Paths.

Context Brief Advanced Features

  • Multi-level context anchoring ranging from transient notes to deep state preservation.
  • Intelligent secret redaction that automatically strips sensitive credentials before writing to disk.
  • Multilingual support that automatically detects and matches the user's current interaction language.
  • Automatic anchor recovery that scans for existing checkpoints within a 48-hour window upon session resumption.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*