perspective-mcp CLAUDE.md

perspective-mcp CLAUDE.md is an instructions file for coding agents from FiveOhhWon/perspective-mcp. It costs 474 tokens per session, scanned A, original, MIT.

Project instructions for building and testing a TypeScript MCP server, a program that makes analysis tools available to AI agents and other clients.

In plain words
What is it for?
They help develop the server, compile it, run it in development or production, test it directly, and work with its sequential-analysis and structured-debate modes.
Why use it?
They give contributors the commands and architectural context needed to run the server, understand its parts, and test changes.

Instructions file

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/fiveohhwon/perspective-mcp/claude-md
Clone the repo
git clone --depth 1 https://github.com/FiveOhhWon/perspective-mcp

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for perspective-mcp CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/fiveohhwon/perspective-mcp/claude-md.svg)](https://agentmods.dev/instructions/fiveohhwon/perspective-mcp/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/fiveohhwon/perspective-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/fiveohhwon/perspective-mcp/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 474 This file is loaded in full into every session.
When invoked 474 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00474 $0.00474
Opus 5 $0.00237 $0.00237
Sonnet 5 $0.00095 $0.00095
Haiku 4.5 $0.00047 $0.00047

Measured 4d ago against content hash 88df639f738b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

perspective-mcp CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 60 lines

What it actually says

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Commands

Build and Development

npm run build    # Compile TypeScript to JavaScript (outputs to dist/)
npm run dev      # Run in development mode with hot reload using tsx
npm start        # Run the compiled production version

Testing the MCP Server

# Run the server directly for testing
npx tsx src/index.ts

# Or after building
node dist/index.js

Architecture

This is an MCP (Model Context Protocol) server implemented as a single TypeScript file (src/index.ts) that provides tools for multi-perspective analysis and structured debates.

Core Components

  1. PerspectiveAnalysisServer Class: Manages all state for both analysis modes

    • perspectives: Array of professional roles with focus areas
    • analyses: Map tracking analysis for each perspective
    • debateState: Manages ongoing debate sessions
  2. Two Operation Modes:

    • Sequential Analysis: Each perspective analyzes in order, building on previous insights
    • Debate Mode: 2-4 perspectives engage in structured 3-round debates
  3. MCP Tools Exposed:

    • set_perspectives: Initialize with professional roles
    • perspective: Perform analysis from current perspective
    • start_debate: Begin structured debate
    • debate_turn: Submit debate statement
    • inject_constraint: Add constraints mid-debate
    • debate_summary: Generate final summary

Key Implementation Details

  • Uses stdio transport for MCP communication
  • Strict TypeScript with comprehensive type definitions
  • Error handling with MCP-specific error codes (e.g., -32602 for invalid params)
  • State validation ensures proper workflow progression
  • All server state is managed in-memory (resets on restart)

Development Notes

  • The server is stateful - perspectives must be set before analysis
  • Debate mode enforces strict turn order and round progression
  • Each debate round has specific purposes (initial positions → responses → synthesis)
  • The nextPerspectiveNeeded flag controls analysis flow
  • Constraint injection only works during active debates
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 60 lines · 474 tokens per session scan A 88df639f738b

Subscribe to this mod's changes

perspective-mcp CLAUDE.md is an instructions file published in the GitHub repository FiveOhhWon/perspective-mcp (2 stars, last pushed 6mo ago), licensed MIT. It adds 474 tokens to every session, about $0.0024 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.