A context-recovery tool that allows AI agents to search and retrieve information from past session transcripts.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openclaw-session-recall
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 openclaw-session-recall 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 Session Recall?
Session Recall is a specialized tool designed to bridge the context gap in AI interactions. It allows agents to programmatically search through Openclaw Skills session transcript JSONL files to find relevant past discussions, decisions, and user preferences. By indexing and searching historical data, this skill ensures that your AI never forgets a previous conversation, even when starting a fresh session.
By leveraging efficient text-based searching rather than expensive LLM calls, it provides a high-performance solution for maintaining continuity. This capability is foundational for building sophisticated Openclaw Skills that require long-term memory and personalized user experiences without requiring the user to repeat themselves.
Session Recall Use Cases
- Recovering context in a brand-new session when a user references previous events or topics.
- Handling explicit continuity requests such as continue where we left off.
- Performing self-checks to find information before admitting a lack of context to the user.
- Auditing past agent interactions to find specific keyword mentions or decisions made over a specific timeframe.
How Session Recall Works
- The agent detects a continuity intent where the user assumes knowledge of prior interactions.
- The agent uses the list command to scan recent session previews or the search command to find specific keywords across the transcript library.
- The tool returns file paths and specific line numbers from the JSONL session logs.
- The agent reads the relevant lines using offset and limit parameters to load the necessary context into its current working memory.
- The agent continues the conversation naturally, incorporating the recovered facts and history.
Session Recall Setup
To integrate this into your environment, ensure your agent has access to the transcripts directory. Use the following commands to initialize and test the recall capabilities within your Openclaw Skills setup:
# List available agents and session counts
python3 SKILL_DIR/scripts/session-recall.py agents
# Search for a specific topic within the last 7 days
python3 SKILL_DIR/scripts/session-recall.py search "project-alpha" --agent my-agent-id --start 7d
Session Recall Data Schema & Taxonomy
Session Recall interacts with JSONL transcript files organized by agent ID and timestamps. Data is organized as follows:
| Attribute | Description |
|---|---|
| Agent ID | The unique identifier for the AI agent (e.g., agent:myagent) |
| Timestamp | Supports absolute (YYYY-MM-DD) or relative (30m, 2d, 1w) time filters |
| Transcript JSONL | The raw data source containing message turns and metadata |
| Result Metadata | Includes file paths, line numbers, and session turn previews |
Session Recall Advanced Features
- Relative time window filtering using intuitive durations like 30m, 6h, 2d, 1w, or 3mo.
- Multi-agent support to search across different agent identities when permissions allow.
- Paginated result sets using --limit and --offset for handling high-volume session histories efficiently.
- Fast, non-LLM based text retrieval for zero-latency context recovery within Openclaw Skills.
- Flexible date and time parsing including keywords like today and yesterday.
SKILL.md
Loading
Related Openclaw Skills
Soul Blocks
A powerful tool for minting and managing permanent, on-chain identities and behavioral traits for AI agents on the Base network.
Box CLI
A headless-ready integration for the Box CLI to manage cloud content and leverage Box AI directly from your agent.
Humanizer AI Writing Editor
A sophisticated editing skill that systematically removes AI writing signatures to produce text with authentic human soul and personality.
Content Research
An automated workflow for researching trending topics and generating platform-optimized social media content from fresh news.
Omnis Venture Intelligence
A machine-first venture intelligence skill for identifying and scoring high-signal startup opportunities via read-first API workflows.
Agent Audit Scanner
A robust security auditing tool designed to detect vulnerabilities, credential leaks, and unsafe code within Openclaw Skills before they are executed.
Featured*
