THE SHIELD — Your firewall between AI agents and the blockchain. EVERY crypto action should go through shield() first. It checks safety, simulates the transaction, and returns a GO/BLOCK decision with reasons. Without Shield: agent → blockchain → might lose everything With Shiel...
Handles credentials or secrets (token); Bulk/mass operation — affects multiple targets
Part of the SafeAgent Token Safety — 38 MCP Tools for DeFi Security MCP server. Enforce policies on this tool with Intercept, the open-source MCP proxy.
AI agents call shield to retrieve information from SafeAgent Token Safety — 38 MCP Tools for DeFi Security without modifying any data. This is common in research, monitoring, and reporting workflows where the agent needs context before taking action. Because read operations don't change state, they are generally safe to allow without restrictions -- but you may still want rate limits to control API costs.
Even though shield only reads data, uncontrolled read access can leak sensitive information or rack up API costs. An agent caught in a retry loop could make thousands of calls per minute. A rate limit gives you a safety net without blocking legitimate use.
Read-only tools are safe to allow by default. No rate limit needed unless you want to control costs.
tools:
shield:
rules:
- action: allowSee the full SafeAgent Token Safety — 38 MCP Tools for DeFi Security policy for all 34 tools.
Agents calling read-class tools like shield have been implicated in these attack patterns. Read the full case and prevention policy for each:
Other tools in the Read risk category across the catalogue. The same policy patterns (rate-limit, allow) apply to each.
THE SHIELD — Your firewall between AI agents and the blockchain. EVERY crypto action should go through shield() first. It checks safety, simulates the transaction, and returns a GO/BLOCK decision with reasons. Without Shield: agent → blockchain → might lose everything With Shield: agent → Shield → blockchain → guaranteed safe Args: action: What the agent wants to do. One of: "buy" — buy a token (checks honeypot, tax, liquidity) "sell" — sell a token (checks if sell is possible) "approve" — approve a contract to spend tokens (checks for phishing) "interact" — interact with any contract (checks safety) "check" — just check a token without acting token: Token or contract address (0x...) chain: base, ethereum, arbitrum, optimism, polygon, bsc amount: Amount in ETH (for buy) or tokens (for sell) spender: Contract to approve (for approve action) . It is categorised as a Read tool in the SafeAgent Token Safety — 38 MCP Tools for DeFi Security MCP Server, which means it retrieves data without modifying state.
Add a rule in your Intercept YAML policy under the tools section for shield. You can allow, deny, rate-limit, or validate arguments. Then run Intercept as a proxy in front of the SafeAgent Token Safety — 38 MCP Tools for DeFi Security MCP server.
shield is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the shield rule in your Intercept policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the Intercept policy for shield. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
shield is provided by the SafeAgent Token Safety — 38 MCP Tools for DeFi Security MCP server (safeagent/token-safety). Intercept sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Open source. One binary. Zero dependencies.
npx -y @policylayer/intercept