$ show ai-boost/awesome-prompts
The ai-boost/awesome-prompts repository is a curated collection of prompt templates designed to configure AI agents for specific roles and tasks. It contains hundreds of specialized prompts for agent architectures (e.g., personal agent brains, multi-agent orchestrators), agent skills (e.g., coding agents, security auditors), and domain-specific agents (e.g., healthcare operations, legal analysts). Each prompt encapsulates non-obvious design patterns for agent systems, such as three-layer retrieval for memory, plan-execute loops, declarative UIs, and skill-based architectures. The repository serves as a reference for prompt engineering and agent design, enabling users to build reliable, safe, and effective LLM-based agents without starting from scratch.
The LLM identified 5 files worth highlighting for reuse:
Implements a sophisticated agent memory system with three-layer retrieval (exact match -> graph traversal -> vector search) and a self-maintenance cycle (dream cycle) for continuous improvement. This pattern is reusable for any agent requiring long-term knowledge retention and self-improvement.
Defines a declarative UI architecture for AI agents, where agents generate UI descriptions that are rendered by a trusted client. Includes component catalog, data model and bindings, actions and RPC, and streaming updates. This pattern is reusable for building agent-user interfaces.
Describes a plan-then-execute loop for AI agents, emphasizing MCP-native tool discipline, a tiered permission model (read-only, prompt-confirm, auto mode), and session management. This pattern is reusable for creating reliable and safe agent execution loops.
Provides a framework for auditing agent skill ecosystems to detect compositional risks where individually safe skills combine into unsafe capabilities. Includes a capability model across 8 categories and forbidden composition patterns. This pattern is reusable for ensuring safety in modular agent systems.
Details the deployment and operation of a codebase-memory MCP server, including indexing strategy, mapping agent questions to MCP tools, designing query plans, interpreting graph results, and integrating with coding-agent workflows. This pattern is reusable for agents that need to understand and navigate large codebases.