Skills

Enables real-time web search, targeted page retrieval, and structured web crawling for Retrieval-Augmented Generation (RAG).

Extracts entities and semantic relations from unstructured sources to construct a queryable knowledge graph.

Standardizes the ingestion of critical outcomes into lightweight AI notes or structured long-term memory (LTM) episodes.

Initializes the operational environment within the default context, introduces the engine architecture, and facilitates the transition to dedicated workspaces.

Enables task delegation to isolated execution environments, supporting both sequential and parallel subagent lifecycles with automated result aggregation.

Autonomous software engineer agent capable of implementing features, running tests, and debugging within a codebase.

High-precision information extraction from unstructured documents via semantic search, targeted sampling, and sequential traversal.

Reconstructs historical context by querying message chains and long-term memory (LTM).

Manage time-triggered workflows using cron expressions, fixed intervals, or specific timestamps.

Protocol for auditing, repairing, or purging problematic entries within the Long-Term Memory (LTM) cache to maintain data integrity.

Orchestrates the lifecycle of user-defined skills, enabling discovery of engine capabilities and the creation, retrieval, modification, or deletion of custom skill definitions.

Orchestrates the meeting lifecycle, from pre-meeting context analysis to automated capture of decisions and action items.

Automating follow-ups, recurring checks, and delegated background tasks through structured job scheduling and subagent delegation.

Synthesizes multi-source context, including chat history, long-term memory, indexed collections, and documents, into decision-ready briefings.

Efficient querying of embedding collections via semantic similarity search.