Enables running AI agents via OpenAI-compatible APIs with custom system prompts, models, and queries. Supports persistent memory, preset agents, and multi-step workflows like pipelines and swarms.
Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find.
Install from
M8ven verifies MCPs across every public registry — install directly from whichever one you prefer.
Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.
When Vitest UI server is listening, arbitrary file can be read and executed
@modelcontextprotocol/sdk has cross-client data leak via shared server/transport instance reuse
Anthropic's MCP TypeScript SDK has a ReDoS vulnerability
Model Context Protocol (MCP) TypeScript SDK does not enable DNS rebinding protection by default
yaml is vulnerable to Stack Overflow via deeply nested YAML collections
process.env. You'll be asked to provide them before it can run.AGENTS_MCP_DATA_DIR— ~/.agents-mcp Directory for the SQLite database (agents-mcp.db).AGENT_ALLOW_CUSTOM_BASE_URL— true Set to false to prevent callers from overriding the base URL (SSRF guard).AGENT_API_BASE— "": "http://127.0.0.1:3030/v1",AGENT_API_KEY— "": "optional",AGENT_CONTEXT_WINDOW— 128000 Token budget for context window management.AGENT_EMBEDDING_BASE— This feature requires the endpoint to support the /embeddings route. If it doesn't, embedding silently fails and agents continue without semantic recall.AGENT_EMBEDDING_KEY— Falls back to AGENT_API_KEY API key for the embedding endpoint.AGENT_EMBEDDING_MODEL— text-embedding-3-small Embedding model for semantic vector memory.AGENT_MAX_CONCURRENT— 10 Maximum number of concurrent in-flight API requests.AGENT_MODEL— "": "gpt-5.3-codex"AGENT_TIMEOUT_MS— 300000 Request timeout in milliseconds.PRESETS_DIR— ~/.agents-mcp/presets Directory where agent presets are stored.[](https://m8ven.ai/mcp/dedinc-agents-mcp-1hlaq0)