What problem does it solve? Running the last30days research engine inside the Exocórtex ecosystem involves provider configuration, API key isolation, silent credit failures, and thin results for B2B or regional topics. This Skill consolidates the operational knowledge needed to install, configure, diagnose, and troubleshoot the engine so research pipelines return reliable multi-source data. ## Core Features & Use Cases - Provider Configuration: Seeds the engine's reasoning and vision models from central Exocórtex LLM roles, with a DeepSeek patch via lazy URL/model resolution in providers.py. - Failure Diagnostics: Ships a pre-flight probe script that classifies ScrapeCreators API keys as healthy, one-shot, dead, bad-key, or upstream-down before wasting a 60-180s pipeline run. - Fallback Pipelines: Supplies Google News RSS parsing, competitive digital maturity probing, and zero-signal synthesis patterns for B2B and regional Brazilian topics where social sources return nothing. - Use Case: An executive asks for TikTok and Reddit trends on a topic. You run the SC key diagnostic first, confirm the key is healthy, execute the engine with a query plan, and supplement thin results with Google News RSS. ## Quick Start Run the last30days engine to research a topic across Reddit and YouTube from the last 30 days after verifying the ScrapeCreators key with the diagnostic script.
