skills.rest
augustscl
augustsclCOMMUNITY
Explore all skills by this author →
GitHub

memory-deposit

Verify and repair workspace memory directories, scripts, and vector embeddings.

2|Updated Apr 6, 2026
Search Tags:#automation#git#memory#merge#transcripts#embedding#vector-search
One-click install
npx skills add https://github.com/augustscl/awesome-xiawang-skills --skill memory-deposit-augustscl
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: memory-deposit
Source: https://github.com/augustscl/awesome-xiawang-skills/tree/main/memory-deposit
Command: npx skills add https://github.com/augustscl/awesome-xiawang-skills --skill memory-deposit-augustscl

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Detects and repair gaps in a workspace's multi-layer memory system by verifying required directories, merge and commit scripts, heartbeat and agent configuration, and vector recall readiness so agents stop forgetting or failing to persist important context.

Core Features & Use Cases

  • Comprehensive health check: Verifies presence of memory/transcripts, memory/projects, scripts/merge-daily-transcript.js, scripts/auto-commit.sh, .git, HEARTBEAT.md entries, and AGENTS.md memory rules.
  • Automated remediation guidance: Copies or instructs placement of required scripts, scaffolds directories, patches HEARTBEAT.md and AGENTS.md from references, and enforces voice JSONL role conventions.
  • Vector recall validation: Runs memory_search checks and diagnoses missing embedding keys or lack of content, guiding the user to configure an embedding provider.
  • Operational use cases: Initializing memory after install, diagnosing why an agent "forgets", ensuring daily transcript merging and safe auto-commits, and preparing memory for reliable retrieval and project save/load workflows.

Quick Start

Run a full memory deposit check to validate, repair, and report the workspace memory layers and vector search readiness.

Frequently Asked Questions about memory-deposit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
Why does my AI agent forget context and fail to persist important workspace information?▼

Agent memory loss occurs when workspace layers like transcripts, project memory, and vectorized long-term memories are incomplete. A memory health check diagnoses missing directories, broken merge scripts, and unconfigured vector embeddings to repair context persistence.

How do I set up and validate a multi-layer workspace memory system with transcripts and vector search?▼

To validate workspace memory, scaffold directories like memory/transcripts and memory/projects, copy merge-daily-transcript.js and auto-commit.sh scripts, patch HEARTBEAT.md and AGENTS.md, and verify vector embedding readiness for memory_search.

What is vector recall validation and how does it work for agent memory search?▼

Vector recall validation runs memory_search checks to diagnose missing embedding keys or lack of content, guiding you to configure an embedding provider. This ensures vectorized long-term memories are ready for reliable retrieval during agent workflows.

Can I use git auto-commit scripts and daily transcript merging for workspace memory backup?▼

Yes, the memory system supports safe auto-commits and daily transcript merging by verifying and placing required scripts like auto-commit.sh and merge-daily-transcript.js. This ensures transcript layers are consistently backed up and merged.

What are the limitations when repairing inconsistent memory behavior in an OpenClaw workspace?▼

Repairing inconsistent memory behavior requires filesystem access to verify directories, scripts, and configuration files like AGENTS.md and HEARTBEAT.md. Vector recall validation also depends on having a properly configured external embedding provider.

Do I need to configure an embedding provider before running a memory deposit check?▼

You can run a memory deposit check without an embedding provider, but vector recall validation will diagnose missing embedding keys and guide you to configure one. Full memory search functionality requires a configured embedding provider for vectorized content.

Related Skills

Explore More Software Engineering Skills→
NominexHQNominexHQ
15

pmm-init

Initialize structured git-backed memory with guided configuration and scaffolding.

Official
Advanced
wewetv1987-cellwewetv1987-cell

openclaw-mem

Preserve knowledge across OpenClaw session compaction with MEMORY.md and daily logs.

Community
Advanced
memorysavermemorysaver

project-behavior

Configure project-level agent behavior by editing AGENTS.md files.

Community
Intermediate
RamachandrajoshiRamachandrajoshi

using-git-worktrees

Create isolated Git worktrees with safety checks and automated setup.

Community
Advanced