191 prompts
Multi-step prompts for planning, tool choice, state tracking, and action boundaries.
An advanced prompt framework designed to drive GPT models in automatically executing complex tasks, supporting step-by-step execution, tool invocation, and structured output. Ideal for large-scale project development, content creation, and data analysis.
A long-horizon research and task-completion agent that navigates the web, extracts structured information, and executes multi-step workflows with disciplined tool use and explicit reasoning.
Designs product systems where AI is the foundational layer, not a feature—emphasizing agentic workflows, generative interfaces, and self-improving loops.
Package domain expertise into a portable, reusable skill that other agents can load on demand to perform recurring workflows safely and consistently.
An expert coding agent prompt emphasizing planning before coding, security-first practices, test-driven development, and minimal changes for production-ready code generation and modification.
An evidence-based code reasoning specialist system designed to analyze code issues and guide code changes through verifiable reasoning grounded in actual codebase evidence.
Design a comprehensive agile transformation program for enterprise organizations, covering strategic diagnostics, operating model design, framework selection, product management integration, technical practices, metrics, culture change, and capability building. Emphasizes moving beyond agile theater to true adaptability, incorporating AI's impact on team structures, and providing actionable roadmaps with risk mitigation, anti-fragility, and continuous improvement mechanisms.
Design memory systems for long-running agents to learn from experience, avoid repetitive mistakes, and retrieve relevant context at the right time—without token bloat or stale recollections.
Design efficient, minimal, and secure tool suites for agents, adhering to the 'tool-as-interface' principle, optimizing namespaces, descriptions, and error handling so agents can accurately discover and use tools.
Audits agent skill ecosystems for supply-chain poisoning, self-propagating attacks, and privilege escalation risks across SKILL.md, MCP servers, tool schemas, and shared memory pools.
Design a safe, debuggable, reversible, and measurable runtime environment for an agent, clarifying the model's role within a larger production system.
Decide how agents, tools, and interfaces should communicate in a production system, focusing on interoperability, safety boundaries, and long-term maintainability.
Design cooperation mechanisms for multi-agent systems that promote collaboration when beneficial while preserving healthy competition and independent verification.
Design modular, progressively loaded skills for ADK-style agents with layered metadata, lifecycle management, and built-in verification.
Define authority, responsibility, and control boundaries across multiple agents to ensure the system remains auditable, safe, and predictable. Prevent uncontrolled delegation by establishing clear ownership, permissions, delegation rules, and approval mechanisms.
Design real-world useful AI agent evaluations that separate model capability, harness quality, tool reliability, and environment noise through executable tasks, safety boundaries, and multi-dimensional scoring.