AI agents use close_i2c to create or update resources in Buspirate — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Buspirate environment.
This tool terminates a session and resets the BusPirate to a safe high-impedance (HiZ) state. It modifies device state reversibly—the session can be reopened by initializing I2C again. This is a safe cleanup operation typical of session-management tools. It does not retrieve data (Read), execute arbitrary code (Execute), delete irreversibly (Destructive), or move money (Financial).
From the tool's definition close_i2c: 'Close an I2C session and reset the BusPirate to HiZ mode.' The tool resets hardware state, explicitly marked [allowed-write], with instant duration.
Attacks that exploit this kind of access
Close an I2C session and reset the BusPirate to HiZ mode. [allowed-write] [Duration: instant.]. It is categorised as a Write tool in the Buspirate MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Buspirate MCP server in PolicyLayer and add a rule for close_i2c: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Buspirate. Nothing to install.
close_i2c is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the close_i2c rule in your PolicyLayer 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 PolicyLayer policy for close_i2c. 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.
close_i2c is provided by the Buspirate MCP server (mplogas/buspirate-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →