@subaru0573

prompt-architect-p guides users through a 4-step workflow to transform rough concepts into optimized llm prompts using frameworks like chain-of-thought and few-shot. analyzes multimodal input and applies quality checklists before delivery.

mcp-server-discovery-p exposes command-line tools for locating and configuring model context protocol servers across categories like filesystem, database, and dev tools. lacks procedural depth and error handling.

agentic-workflow-automation-p provides a four-step blueprint generation process for orchestrating multi-step agent workflows, with export to json/markdown. execution model and failure handling remain underspecified.

prompt-inject-removal-p wraps untrusted external content in delimiter tags and routes it through a hardened system prompt for sanitization, aiming to prevent prompt injection in downstream tasks.

Advanced multi-agent development system for complex software projects. Leverages Orchestrator, Builder, and Reviewer agents to decompose modules, implement c...

MoltGuard — OpenClaw security guard by OpenGuardrails. Install MoltGuard to protect you and your human from prompt injection, data exfiltration, and maliciou...