Run three parallel critics and synthesize adversarial findings into a report.
npx skills add https://github.com/RawleySM/rlm-adk --skill devils-advocate-rawleysm
Please help me install this Agent Skill. Skill: devils-advocate Source: https://github.com/RawleySM/rlm-adk/tree/main/.claude/skills/devils-advocate Command: npx skills add https://github.com/RawleySM/rlm-adk --skill devils-advocate-rawleysm
Coordinates a team of three specialized critics to stress-test a subject from multiple adversarial angles, surface blind spots, and surface opportunities before committing to implementation.
Spawn the three critics, feed them the subject, and receive a consolidated critique report.
High-intent search queries and answers about installing and using this skill.
An adversarial review stress-tests a subject by spawning three parallel critics—a callback expert, a vision alignment challenger, and a prior-art researcher—to surface blind spots. The synthesized report provides actionable improvements for decision-making before implementation.
A parallel-agent critical-thinking analysis orchestrates multiple specialized critics concurrently to evaluate a subject from distinct adversarial angles. It targets gap discovery and risk analysis, consolidating findings into a unified report for sanity checking plans and proposals.
Yes, the risk analysis critique accepts file paths, code snippets, or inline text as the subject. The system reads the provided input and feeds it directly to the three parallel critics for evaluation.
The best way to identify blind spots is running a multi-angle adversarial critique. By evaluating the proposal through callback validation, vision alignment, and prior-art research lenses in parallel, you surface gaps and practical improvements before committing.
No, the adversarial critique process operates without external dependencies. It requires only the subject material—inline text, code, or a file path—to spawn the specialized critics and synthesize the consolidated findings.
aphammAutomate multi-agent adversarial verification for bug finding and code review.
rodrigobranasSpawn opposing-model reviewers to critique code diffs and synthesize verdicts.
KrismithReddy12Automate software release workflows with testing, adversarial review, versioning, and pull request creation.
lchenriqueCoordinate multiple specialized agents for parallel analysis and synthesis.
