openclaw-memory-stackTotal recall, 90% fewer tokens. Open source OpenClaw memory plugin — 5-engine local search, structured fact extraction, smart dedup, cross-agent sharing, and...
Install via ClawdBot CLI:
clawdbot install apptah/openclaw-memory-stackGrade Good — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Potentially destructive shell commands in tool definitions
curl -fsSL https://bun.sh/install | bashCalls external URL not in known-safe list
https://openclaw-memory.apptah.comUses known external API (expected, informational)
api.openai.comAI Analysis
The skill's primary data processing (5 search engines, fact extraction) is local, and its external network calls are documented for license activation/verification and optional cloud LLM fallback. The 'UNSAFE_SHELL' signal for installing 'bun' is a standard installation command for a required dependency, not inherently malicious. The risk is low as no evidence suggests unauthorized data exfiltration or hidden malicious instructions.
Generated Aug 6, 2026
A customer support agent uses the memory stack to instantly recall past interactions, preferences, and issue resolutions, reducing response time and improving accuracy. It avoids re-reading full conversation histories, cutting token usage by 90%.
A research assistant AI uses the memory stack to store and retrieve summaries, key findings, and citations from thousands of papers. The vector search and fact extraction enable quick retrieval of relevant information without re-processing full texts.
A personal AI assistant uses the memory stack to remember user preferences, personal details, and past conversations, providing personalized responses and proactive suggestions. The local storage ensures privacy and offline functionality.
A sales AI uses the memory stack to track client interactions, deal progress, and stakeholder preferences, enabling tailored follow-ups and better forecasting. The entity tracking and dedup features maintain a clean, up-to-date CRM.
A coding assistant uses the memory stack to remember code context, past refactors, and common patterns across projects. Code search capability allows quick retrieval of relevant snippets, reducing token-heavy context re-reading.
The core plugin is free and MIT-licensed, but companies can pay for enterprise support, customization, and training. Revenue comes from support contracts and SLAs.
Offer the basic memory stack for free, with premium tiers that include advanced features like enhanced fact extraction, dedicated LLM integration, or priority updates. Revenue from subscription tiers.
License the memory stack as a white-label solution to other AI agent platforms, integrating it into their products. Revenue through licensing fees or per-seat royalties.
💬 Integration Tip
Start by swapping in the memory stack for default memory in a test environment, leveraging its local storage and search to see immediate token savings. Integrate with your existing LLM endpoint via the OPENCLAW_LLM_API_KEY for fact extraction.
Scored Aug 6, 2026
Audited Apr 16, 2026 · audit v1.0
Structured reasoning through sequential thinking — break complex problems into steps, solve each independently, verify consistency, synthesize conclusions wi...
Loads and manages company context for all C-suite advisor skills. Reads ~/.claude/company-context.md, detects stale context (>90 days), enriches context duri...
Automatically prune and compact agent memory files to prevent unbounded growth. Circular buffer for logs, importance-based retention for state, and configura...
Founder onboarding interview that captures company context across 7 dimensions. Invoke with /cs:setup for initial interview or /cs:update for quarterly refre...
Ultimate AI agent memory system for Cursor, Claude, ChatGPT & Copilot. WAL protocol + vector search + git-notes + cloud backup. Never lose context again. Vib...
Expert skill for memory-lancedb-pro — a production-grade LanceDB-backed long-term memory plugin for OpenClaw agents with hybrid retrieval, cross-encoder rera...