Orchestrates parallel retrieval of distinct temporal or spatial data points followed by a centralized calculation and verification step.
Orchestrates a multi-stage filter-and-verify process to find a specific entity satisfying three or more independent constraints.
Decomposes queries where each answer serves as the mandatory search key for the subsequent step in a sequential chain.
A two-stage strategy to isolate specific visual or bibliographic details from social media and video content by separating source identification from contextual extraction.
Identifies the task type and directs the orchestrator to the correct decompose skill.
Analyze local images using an OpenAI-compatible multimodal endpoint. Use when the user question depends on visual content from a local image file — visual question answering, describing images, reading text in images, identifying objects, etc.
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
Install and manage Python packages using uv pip. Use when a Python import fails with ModuleNotFoundError, user asks to install a package, or a script requires a missing dependency.