Skill UI

This tool provides state-of-the-art semantic memory search across various knowledge domains. It supports multiple advanced retrieval strategies, including hybrid (sparse+dense) searching and Graph RAG for multi-hop reasoning. Key features include Mean Margin Ranking (MMR) for result diversity, recency weighting for timely information, and structured namespace filtering, ensuring highly relevant and context-aware results.

Use the Monitor tool to achieve real-time observability of swarm events instead of relying on periodic polling. This utility streams comprehensive NDJSON events, including agent spawns, task completions, memory writes, and health checks. It is designed for developers and engineers needing immediate insight into the operational status and health of a distributed system.

This tool synthesizes deep, accumulated research findings stored in memory into structured, actionable reports. It systematically grades evidence quality (High, Medium, Low), identifies and resolves conflicting claims, and generates comprehensive outputs including an executive summary, ranked key findings, limitations, and concrete recommendations. Ideal for transforming raw research data into strategic insights.

This utility diagnoses and automatically repairs common issues within the Ruflo installation environment. It performs comprehensive health checks on core dependencies like Node.js, npm, and git, while also validating configurations, daemon status, memory databases, and API keys. It provides targeted commands to resolve issues related to memory initialization, daemon startup, and configuration resets, ensuring a stable operational environment.

This tool manages agent memory stored in the RVF (Ruflo Vector Format). It facilitates the backup, transfer, and sharing of complex agent knowledge, including vector embeddings, metadata, and causal relationships, across different projects and platforms. Users can list, export, import, migrate, and analyze memory statistics to ensure knowledge continuity.

A comprehensive tool for managing and checkpointing agent sessions. It allows users to save the entire operational state, including memory, configuration, and task progress, across multiple conversations. Essential for long-running tasks, complex workflows, or when needing to pause and resume work seamlessly with full context preservation.

This pattern demonstrates how to build a robust, thread-safe data persistence layer in Swift using the 'actor' model. It combines fast in-memory caching with reliable file-backed storage, automatically eliminating data races and the need for manual locks. Ideal for offline-first applications and managing shared mutable state in modern Swift concurrency.

Tree Ring Memory is a local-first, framework-agnostic layer for managing durable knowledge in AI agents. It provides structured methods for recalling past decisions, recording evidence, auditing project state, and deliberately forgetting information, ensuring the core project memory remains clean and actionable without relying on raw conversation transcripts.

A comprehensive vector search plugin offering two distinct paths: large-scale corpus search using HNSW (supporting RaBitQ quantization for memory efficiency) and a fast WASM-backed router for high-priority hot patterns. Ideal for implementing Retrieval-Augmented Generation (RAG), semantic search across massive datasets, or routing queries to specific, limited patterns.