Multi-agent strategic review of documents, plans, and proposals. Commissions review agents and iterates until the review meets quality standards.
From aops-corenpx claudepluginhub nicsuzor/aopsThis skill is limited to using the following tools:
review-contexts/execution-environments.mdreview-contexts/pr-code.mdreview-contexts/pr-framework.mdDesigns and optimizes AI agent action spaces, tool definitions, observation formats, error recovery, and context for higher task completion rates.
Enables AI agents to execute x402 payments with per-task budgets, spending controls, and non-custodial wallets via MCP tools. Use when agents pay for APIs, services, or other agents.
Compares coding agents like Claude Code and Aider on custom YAML-defined codebase tasks using git worktrees, measuring pass rate, cost, time, and consistency.
Multi-agent strategic review of documents, plans, and proposals. The orchestrator is James — commission James and let him manage the agent loop. If you are James, this skill is your operating context.
Use this when a document needs strategic review, not proofreading:
Commission James as the orchestrator. He manages the agent loop, evaluates output quality, iterates, and synthesises.
Agent(subagent_type="aops-core:james", prompt="[artifact + context]")
James will commission the right agents based on the artifact type and load the appropriate review context descriptor. You do not need to manage the agent loop — James does that.
Context descriptors in review-contexts/ configure James's behavior per artifact type:
| Descriptor | When to use |
|---|---|
pr-code.md | Code PRs — features, fixes, refactors |
pr-framework.md | Framework PRs — skills, agents, hooks, enforcement, workflows |
James will read the relevant descriptor automatically based on what you tell him about the artifact.
| Agent | What they do | Ruth always runs |
|---|---|---|
| rbg | Axiom compliance and workflow discipline — The Judge | Yes — non-negotiable |
| pauli | Strategic critique via 10 cognitive moves — The Logician | As needed |
| marsha | Independent runtime verification — The QA Reviewer | When code is involved |
The loop exists because one-shot prompting reliably produces competent-but-not-genius reviews: internally consistent, surface-level, answering the question as posed. James's job is to force elevation — from instance to class, from artifact to process, from "is this right?" to "is this the right question?". He also carries axiom compliance (Ruth) and runtime verification (Marsha) as non-negotiable dimensions that strategic review alone cannot provide.