A structured AI agent skill designed to capture and extract durable, high-value technical knowledge at the end of a development session for long-term retention.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install memory-handoff
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 memory-handoff 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 Memory Handoff?
Coding sessions frequently die alongside their terminal transcripts, losing critical institutional knowledge in the process. The Memory Handoff skill solves this by generating short, highly structured notes containing durable knowledge—such as architecture decisions, root causes, setup gotchas, workflow changes, and security findings—that a long-lived agent or human teammate can review and file. This ensures valuable contextual insights survive past individual session lifecycles.
While designed to pair seamlessly with the brigade automation system for linting and ingestion, this skill functions perfectly as a standalone protocol. It prevents developers from losing hard-earned insights to ephemeral chat histories by forcing the extraction of verifiable facts over session narratives.
Memory Handoff Use Cases
- Capturing non-obvious root causes and bugfix details after a complex debugging session.
- Documenting unexpected environment setup gotchas or workflow changes for future teammates.
- Recording major architectural decisions, research findings, or user preferences.
- Preserving security findings and reusable command patterns discovered mid-development.
- Initiating an explicit session handoff when triggered by user commands like "hand off" or "save this for the memory system".
How Memory Handoff Works
- Trigger Evaluation: At the end of a session, or upon explicit user prompt, the agent evaluates if durable, long-term technical knowledge was uncovered.
- Directory Selection: The agent checks for an existing config folder like
.brigade/,.claude/memory-handoffs/, or.codex/memory-handoffs/to use as the storage target. - Draft Generation: A structured markdown file is drafted using a precise timestamp-slug naming convention (
YYYY-MM-DD-HHMM-<slug>.md). - Structure Formatting: The file is strictly partitioned into metadata blocks including Type, Summary, Durable Facts, Evidence, and Recommended Memory Actions.
- Validation & Linting: The output is run through the brigade CLI linter (if available) or peer-reviewed to ensure it remains factual, specific, and under the 400-word limit.
Memory Handoff Setup
Target Directories
Ensure your repository contains one of the following directories for the agent to dump handoff notes:
.claude/memory-handoffs/.codex/memory-handoffs/.brigade/(If utilizing advanced automation configurations)
CLI Verification Setup
To automate validation, install the brigade utility and run the built-in linter on your repository:
# Lint the repository for handoff structural errors
brigade handoff lint
# Automatically scaffold a blank template when unsure of target expectations
brigade handoff draft
Memory Handoff Data Schema & Taxonomy
File Naming Convention
Handoff files must adhere to the following strict chronological taxonomy:
YYYY-MM-DD-HHMM-<slug>.md
Required Structure Options
| Section | Allowed / Expected Values | Description |
|---|---|---|
| Type | setup, workflow, bugfix, decision, security, preference, research, project-context |
Categorizes the nature of the durable knowledge. |
| Summary | 2 to 4 sentences | High-level synthesis of what occurred and why it matters. |
| Durable Facts | Bulleted list | Verifiable, absolute facts (no relative dates like "yesterday"). |
| Evidence | files changed, commands run, error strings |
Hard proof supporting the discovered facts. |
| Recommended Action | create-card, update-card, no-card |
Explicit instructions for the destination memory system. |
| Target Card | Descriptive filename (e.g., card-name.md) |
Canonical target file in the parent memory registry. |
| Suggested Content | Markdown string beginning with YAML frontmatter | The exact payload payload to ingest. For no-card, this shifts to target documents. |
Memory Handoff Advanced Features
- Brigade Tooling Integration: Native validation matching the strict schemas required by
brigade handoff lintfor automated data pipelines. - Flexible Destination Routing: Swappable formatting contexts using
create-card/update-cardrules with YAML frontmatter or fallbackTarget documenttrees for unmanaged files. - Human-in-the-Loop Safeguards: Designed as a staging protocol that requests reviewer veto power rather than writing directly to canonical memory stores, preserving system integrity.
SKILL.md
Loading
Related Openclaw Skills
Mise
A design-first workflow skill that ensures all engineering tasks are fully designed, scoped, and user-approved before writing a single line of code.
plate
Plate is a prose hygiene and leak-scrubbing skill that ensures outward-facing text like PRs, READMEs, blogs, and commits is free of identity, system, and infrastructure data leaks before publishing.
Website Development Decision Assistant
An automated web-research and scoring agent that evaluates search trends, competitor landscapes, and monetization viability to generate HTML feasibility reports.
Alibaba Cloud SLS DataAgent
An autonomous AI data analyst that translates natural language queries into processed log insights, charts, and structured reports directly from Alibaba Cloud SLS.
fire
The fire skill executes a written implementation plan task-by-task with strict, disciplined verification and no unauthorized improvisation.
expedite
An AI-driven execution engine that systematically processes code audits and bug backlogs, resolving issues one-by-one with rigorous validation and isolated commits.
Featured*
