analyze-1c-research

Automated 1C task analysis uses a three-agent loop with Executor, Reviewer, and Comparator roles to iteratively evaluate and refine solutions until hitting the target score.

A multi-agent workflow for 1C challenges coordinates three distinct agents to execute tasks, review performance, and compare results against targets for autonomous quality refinement.

Run a background script by pointing the command to a Markdown task file, which triggers the autoscoring loop to monitor iterations and output final scores with report paths.

Yes, you can specify a Markdown task file path to start the 1C task orchestration, which the three-agent loop processes to generate structured outputs for audits.

The best way to structure 1C task reports for audits is using the three-agent orchestration loop, which produces structured outputs including final scores, iteration counts, and log paths.