memory-harnessRuntime-enforced memory harness for OpenClaw. Implements 3-stage recall (session preflight, triggered recall, pre-execution gate) with intent classification,...
Install via ClawdBot CLI:
clawdbot install taka3693/memory-harnessGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Generated Mar 21, 2026
A developer resumes work on an ongoing project like OpenClaw, asking to continue from a previous session. The harness triggers recall to fetch the active project, unresolved issues, and recent decisions, ensuring code changes align with prior constraints and avoiding redundant work.
A support agent takes over a customer case where the user references a prior interaction. The harness detects continuation triggers like 'previous issue' and recalls the session digest with pinned facts and unresolved items, enabling personalized assistance without rehashing history.
A lawyer requests edits to a contract draft, mentioning specific clauses or entities. The pre-execution gate recalls prior constraints and project conventions, ensuring modifications comply with legal standards and avoid conflicts from earlier versions.
A researcher asks for help designing an experiment based on earlier notes in tools like Obsidian. The harness triggers recall when entity names or design requests are detected, providing compressed context from recent high-signal items to inform planning.
A product manager references a known project like ClawHub to plan new features. The harness uses entity detection to recall active project details and unresolved items, supporting architecture proposals while skipping generic Q&A about unrelated topics.
Offer the Memory Harness as a cloud-based service with tiered plans based on usage volume and integration depth. Revenue comes from monthly subscriptions, targeting development and support teams needing reliable context recall without heavy infrastructure.
Sell licenses to large organizations for on-premise deployment, with customization options for specific industries like legal or healthcare. Revenue includes upfront licensing fees and ongoing support contracts, ensuring compliance and scalability.
Provide a free version with basic recall features, monetizing through premium upgrades that offer advanced logging, analytics, and integration with tools like Telegram or Obsidian. Revenue streams from upgrade sales and add-on services.
💬 Integration Tip
Integrate the harness by configuring trigger conditions in the memory_policy JSON to match your domain's entities and continuation phrases, ensuring it runs automatically without manual intervention.
Scored Apr 19, 2026
Structured reasoning through sequential thinking — break complex problems into steps, solve each independently, verify consistency, synthesize conclusions wi...
LLM-driven epistemic reasoning engine. Evaluates claims against evidence, outputs calibrated confidence and structured belief state (VERIFIED/CONTESTED/UNCER...
Loads and manages company context for all C-suite advisor skills. Reads ~/.claude/company-context.md, detects stale context (>90 days), enriches context duri...
Turn scattered local sources into a source-constrained evidence notebook for incident, release, and maintainer decisions.
三级记忆管理系统 (Three-Tier Memory Management)。用于管理 AI 代理的短期、中期、长期记忆。包括：(1) 滑动窗口式短期记忆，(2) 自动摘要生成中期记忆，(3) 向量检索长期记忆 (RAG)。当需要管理对话历史、优化上下文、构建个人知识库、或实现记忆持久化时使用此 Skill。
Store secrets, long-term memory, daily logs, and anything custom in your Convex backend instead of local files