ODEI MCP Server
OfficialProvides tools for auditing EVM smart contract code written in Solidity to identify security vulnerabilities using the EVMbench methodology.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@ODEI MCP Server audit this smart contract for security vulnerabilities"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP for ODEI
Public Model Context Protocol interface notes for ODEI integrations.
ODEI is an operational decision engine for individuals: a local-first personal operating layer, semantic world model, and agent control surface designed around user agency.
Public Boundary
This repository is intentionally minimal. It exists to publish the project surface, security contact, and integration direction without exposing production infrastructure or private runtime code.
Published here:
High-level project context
Public interface notes
Security reporting instructions
Non-operational examples or documentation
Not published here:
Production credentials or environment files
Deployment automation for live infrastructure
Private worker, VPN, oracle, or operator code
Internal keys, tokens, state, or telemetry
Related MCP server: Mnemom
Contact
Security reports: security@odei.ai
Product: https://odei.ai
Available Tools
4 toolsodei_guardrail_checkA
Validate an agent action against ODEI's constitutional guardrails. Returns APPROVED, DENIED, or NEEDS_REVIEW with reasoning. Use this before executing any action that could affect finances, reputation, health, or relationships. The guardrail system checks against 7 layers of safety constraints including financial limits, time-of-day restrictions, and value alignment.
Inputs: action (required), context, domain, severity.
odei_smart_contract_auditA
Analyze an EVM smart contract address for security risks. Checks contract verification status, known vulnerability patterns, token safety signals, and cross-references with ODEI's knowledge graph for known entities. Returns a risk score and actionable findings. Supports Base and Ethereum contracts.
Inputs: address (required), chain, check_holders.
odei_world_model_queryA
Search ODEI's constitutional world model — a 10,665-node knowledge graph organized into 6 domains (Foundation, Vision, Strategy, Tactics, Execution, Track). Returns nodes with their types, domains, relationships, and metadata. Use this to understand the structure of any domain, find specific entities, or explore how goals connect to execution.
Inputs: query, domain, type, limit.
odei_world_model_signalA
Rapid trust and relevance scoring for a claim, entity, or topic. Returns a confidence score (0-1), relevance to ODEI's world model, and supporting/contradicting evidence from the knowledge graph. Use this for quick validation before making decisions — it's faster and cheaper than a full world model query.
Inputs: claim (required), category.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 4 tools
Each tool has a clearly distinct purpose: guardrail validation, smart contract auditing, world model querying, and trust scoring. There is no overlap in functionality, and the descriptions clearly differentiate their use cases, making misselection unlikely.
All tool names follow a consistent 'odei_' prefix with descriptive snake_case suffixes (e.g., guardrail_check, smart_contract_audit). This pattern is uniform across all four tools, enhancing predictability and readability.
Four tools is a reasonable count for an MCP server focused on AI safety and blockchain security, but it feels slightly thin given the broad scope implied by the descriptions. Each tool earns its place, but additional tools might be expected for more comprehensive coverage in these domains.
The tools cover key areas like guardrails, smart contracts, and world model interactions, but there are notable gaps. For example, there are no tools for updating or managing the world model, executing actions post-guardrail approval, or handling non-EVM blockchains, which could limit agent workflows in this domain.
Related MCP Connectors
AI agent security, MCP 2026 schema & graph audits, <250ms guardrails, and compliance dossiers
Read-only MCP tools for AI agent discovery, structured resources, and NIULAI information.
Find, vet, and run MCP tools through a secure audited gateway with prompt-injection risk scoring
Your org's AI agents, tasks, runs, search, and brain files as MCP tools and resources.
Related MCP Servers
AlicenseAqualityCmaintenanceExposes TELOS governance primitives—action scoring, receipt verification, Purpose Anchor inspection, audit-chain queries, and CCRS counterfactual replay—as MCP tools, resources, and prompts for any MCP-compatible client.5Apache 2.0- FlicenseNot gradedqualityCmaintenanceProvides trust infrastructure for AI agents by enabling reputation lookup, website trust scanning, and identity verification via MCP tools.1-
- AlicenseAqualityCmaintenanceProvides advisory AI-governance checks for MCP hosts and coding agents, enabling policy verdicts, PII masking, and audit traces before executing tool calls.382 npmMIT

evav-gatewayofficial
AlicenseNot gradedqualityDmaintenanceGoverned MCP gateway that lets AI agents call tools with policy enforcement, prompt-injection screening, a kill-switch, and tamper-evident signed audit logs.Apache 2.0