Server data from the Official MCP Registry
Configure AI agents and diagnose oscillation, overload, freeze, and environment mismatch.
Configure AI agents and diagnose oscillation, overload, freeze, and environment mismatch.
Remote endpoints: streamable-http: https://rpcs1.dev/mcp
This is an MCP server wrapper for the RPCS-1 agent tuner, a configuration framework for AI agents. The server is read-only, deterministic, and focused on parameter recommendations. Code quality is good with proper TypeScript typing and modular structure. The server uses standard production controls (rate limiting, request size limits, host allowlist) and has no malicious patterns, hardcoded secrets, or dangerous operations. Minor findings relate to build artifacts inclusion and dev-time tooling. Supply chain analysis found 1 known vulnerability in dependencies.
4 files analyzed · 3 issues found
Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.
This plugin requests these system permissions. Most are normal for its category.
Available as Local & Remote
This plugin can run on your machine or connect to a hosted endpoint. during install.
From the project's GitHub README.
Configure AI agents that don't oscillate, overload, or freeze.
A configuration framework for AI agents that translates environmental characteristics (entropy, stakes, predictability) into specific LLM parameter recommendations — grounded in RPCS-1 receiver dynamics.
rpcs1-sdk/
├── packages/core/ # TypeScript recommendation engine (@rpcs1/core)
├── sdk/python/ # Python SDK (pip install rpcs1)
└── .github/workflows/ # CI/CD
pip install rpcs1
from rpcs1 import recommend_params
config = recommend_params(
task_description="Customer support agent",
environment_entropy="dynamic",
environment_predictability="somewhat_predictable",
stakes="high",
target_platform="anthropic",
)
print(config.platform_parameters.temperature) # e.g. 0.52
print(config.predicted_regime) # 'stable'
print(config.reasoning) # cites Matching Principle
import { recommend } from '@rpcs1/core';
const rec = recommend({
task: { task_summary: 'Customer support agent' },
environment: {
entropy: 'dynamic',
predictability: 'somewhat_predictable',
stakes: 'high',
context_relevance: 'medium',
commitment_style: 'cautious',
},
target_platform: 'anthropic',
});
console.log(rec.platform_parameters.temperature);
console.log(rec.predicted_regime);
# Install pnpm
npm install -g pnpm
# Install dependencies
pnpm install
# Build and test TypeScript core
pnpm --filter @rpcs1/core build
pnpm --filter @rpcs1/core test
# Test Python SDK
cd sdk/python
pip install -e ".[dev]"
pytest -v
The SDK implements Pred-09-5 from IMM Paper 9:
Stable receivers in an environment with entropy H satisfy TI ~ 1/H.
High-entropy environments → short attention windows (TI ~ 10). Low-entropy environments → long attention windows (TI ~ 90).
Every parameter recommendation traces back to this principle or the basin stability geometry (oscillation/overload/freeze boundary conditions).
Interactive tuner: https://rpcs1.dev
RPCS-1 is also available as a public, anonymous, read-only MCP server:
https://rpcs1.dev/mcp
It exposes one focused tool:
recommend_agent_configuration — use when designing, tuning, or diagnosing an AI agent
against environmental entropy, predictability, stakes, context horizon, and commitment style.The MCP surface intentionally wraps the existing deterministic recommendation engine. Broader communication, market, and decision-analysis tools should be added only after their scoring contracts are implemented and tested in the core package.
Discovery metadata:
server.jsonProduction controls:
MCP_HOURLY_LIMIT controls per-instance MCP throttling (default: 120 requests per IP/hour).MCP_MAX_BODY_BYTES limits request bodies (default: 65536 bytes).MCP_ALLOWED_HOSTS is a comma-separated production host allowlist./api/health reports deployment and MCP readiness metadata.For globally consistent abuse protection across Vercel instances, configure a Vercel Firewall
rate-limit rule for /mcp. The in-process limiter is defense in depth, not a distributed quota.
MIT
Be the first to review this server!
by Modelcontextprotocol · Developer Tools
Read, search, and manipulate Git repositories programmatically
by Modelcontextprotocol · Developer Tools
Web content fetching and conversion for efficient LLM usage
by Toleno · Developer Tools
Toleno Network MCP Server — Manage your Toleno mining account with Claude AI using natural language.