AI agents call capture_code to retrieve information from Agentline without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool retrieves captured codes (likely 2FA/OTP codes from incoming messages or calls) without modifying data. This is fundamentally a read operation. Severity is medium rather than low because 2FA code capture could enable account compromise if misused by an agent, but confidence is reduced to 0.7 due to empty description requiring inference from context.
From the tool's definition Tool name 'capture_code' combined with sibling tool 'capture_email_code' and server description mentioning 'capture 2FA codes' indicates this retrieves/reads verification codes rather than creating, modifying, or deleting them.
Attacks that exploit this kind of access
capture_code. It is categorised as a Read tool in the Agentline MCP Server, which means it retrieves data without modifying state.
Register the Agentline MCP server in PolicyLayer and add a rule for capture_code: 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 Agentline. Nothing to install.
capture_code 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 capture_code 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 capture_code. 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.
capture_code is provided by the Agentline MCP server (jgottlieb84/agentline-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 →