Setup & Installation
openclaw skills install @boscoeuk/context-anchor
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
npx clawhub install context-anchor
What This Skill Does
A bash script that scans local memory and context files to produce a structured briefing about where work left off. Designed for use after context compaction or at session start when working memory is gone.
Because it reads files that survive compaction rather than relying on in-context memory, orientation is consistent regardless of how much context was lost.
When to use it
- Resuming an agent session after context compaction
- Orienting a fresh agent on an in-progress project
- Reviewing decisions made across recent daily logs
- Handing off current state to another agent
- Checking open loops before starting new work
Example Workflow
Here's how your AI assistant might use this skill in practice.
INPUT
User asks: 'What was I working on before the session reset?'
AGENT
- 1Run ./scripts/anchor.sh for a full briefing
- 2Read memory/current-task.md for task status, blockers, and next steps
- 3Scan recent daily log files (last 2 days by default) for decisions and completed items
- 4Check context/active/ for any in-progress task files
- 5Surface open loops marked with TODO, Blocker, or question marks
OUTPUT
A structured briefing showing current task, active context files, recent decisions, and unresolved open loops