AI agents use generate_docs to create or update resources in Tok — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Tok environment.
This tool creates new files with architectural documentation. While file creation is reversible and doesn't destroy data, it modifies the codebase/project state by introducing new documentation files that could affect AI tool behavior and project structure. The generated .ai-context.yaml file is particularly relevant as it guides AI coding tools' decisions.
From the tool's definition Tool generates and creates structured documentation files (ARCHITECTURE.md and .ai-context.yaml). The description states it '생성합니다' (generates/creates) these documents, which are persistent file artifacts.
Attacks that exploit this kind of access
아키텍처 설계 결과를 바탕으로 ARCHITECTURE.md (사람용)와 .ai-context.yaml (AI용) 문서를 생성합니다. It is categorised as a Write tool in the Tok MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Tok MCP server in PolicyLayer and add a rule for generate_docs: 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 Tok. Nothing to install.
generate_docs 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 generate_docs 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 generate_docs. 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.
generate_docs is provided by the Tok MCP server (seongminjaden/tok_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 →