AgentStack
by ssdeanx
Overview
AgentStack is a production-grade multi-agent framework built on Mastra, offering enterprise tools, specialized agents, workflows, and agent networks for scalable AI systems, with a focus on financial intelligence, RAG pipelines, observability, secure governance, and AI chat interfaces.
Installation
npm run devEnvironment Variables
- GITHUB_API_KEY
- GITHUB_PERSONAL_ACCESS_TOKEN
- ALPHA_VANTAGE_API_KEY
- SERPAPI_API_KEY
- POLYGON_API_KEY
- FINNHUB_API_KEY
- POSTGRES_URL
- LANGFUSE_PUBLIC_KEY
- LANGFUSE_SECRET_KEY
- LANGFUSE_HOST
- GOOGLE_API_KEY
- SMITHERY_API_KEY
- SMITHERY_PROFILE
- NEO4J_PASSWORD
- NEO4J_USERNAME
- NEO4J_URI
- NEO4J_DATABASE
- KLAVIS_INSTANCE_ID
- KLAVIS_STRATA
Security Notes
The `execaTool` directly exposes arbitrary command execution to agents, which is a critical security vulnerability allowing a compromised agent to run any shell command on the host machine. The `calendar-tool.ts` uses `execSync` to run `osascript`, a risky primitive if inputs are not strictly sanitized, though current usage appears to involve controlled inputs. `web-scraper-tool.ts` employs robust sanitization (JSDOM/Cheerio) for HTML content and path validation for file writes, which is positive for preventing XSS and path traversal. However, the core issue of arbitrary command execution significantly lowers the safety score.
Similar Servers
MaxKB
An enterprise-grade intelligent agent platform for building knowledge bases, RAG, complex workflows, and AI agents, targeting intelligent customer service and office assistants.
finance-trading-ai-agents-mcp
A specialized MCP server for financial analysis and quantitative trading, designed to deploy local financial MCP services with a departmental architecture for LLM integration and algorithmic trading.
AgentUp
AgentUp is an infrastructure framework for developing, deploying, and managing production-ready AI agents, providing Docker-like consistency, security, and extensibility.
AgentUp
A developer-first framework for building, deploying, and managing secure, scalable, and configurable AI agents, supporting various agent types (reactive, iterative) and the Model-Context Protocol (MCP) for seamless interactions.