Agent Skills · AI Agent
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
A unified entry point for clinical trial design, compliance, QC, and clarification, combining local knowledge with remote Coze analysis and tool orchestration.
Extract reusable lessons from Agent task runs into structured methodology cards, with relation-based recommendations, retrieval, and export.
An OpenClaw auto-updater with dual-source updates, pre-update backups, detailed reports, and scheduled morning digest delivery.
Creates a local three-agent development team in OpenClaw with project, full-stack dev, and testing roles.
Silently recalls memory files, recent logs, and conversation history at session start to preserve user preferences and project context.
Log agent errors, corrections, and feature requests to Markdown, then promote reusable learnings into project memory.
Use durable Markdown files as agent working memory, with task_plan, findings, and progress to reduce context loss.
Adds local memory, workspace seeding, and workflow capture to OpenClaw for cross-session self-correction and maintenance.
A persistent memory tool for AI agents that supports init, encrypted upload, disaster recovery, local and cloud search, watch sync, and distillation hook sync.
A token-efficient task router for general-purpose agents that classifies requests into Ask, Plan, Craft, or Expert modes and applies budget and safety controls.
Generate expert roles, collaboration mechanisms, and prompts for a task while noting that AI usually handles complex work end-to-end without human-style teams.
A memory and knowledge management system for AI agents, covering memory types, gating, retrieval, maintenance, evaluation, and open-source options.