从图片中提取文本内容 Args: image_path: 图片文件路径 language: OCR识别语言，默认为中文简体+英文 Returns: 包含识别结果的字典
AI agents call ocr_recognize_text to retrieve information from Office Editor without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs OCR on an image file and returns the extracted text. It is a read/query operation with no side effects — it reads an image and returns data without modifying or deleting anything.
From the tool's definition 从图片中提取文本内容 (extracts text content from images); takes image_path as input and returns recognition results
Documented attack patterns abuse exactly the kind of access ocr_recognize_text gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Office Editor, and nothing reaches the server without passing your rules. This is the rule we recommend for ocr_recognize_text:
{
"version": "1",
"default": "deny",
"tools": {
"ocr_recognize_text": {}
}
}ocr_recognize_text is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
从图片中提取文本内容 Args: image_path: 图片文件路径 language: OCR识别语言，默认为中文简体+英文 Returns: 包含识别结果的字典. It is categorised as a Read tool in the Office Editor MCP Server, which means it retrieves data without modifying state.
Register the Office Editor MCP server in PolicyLayer and add a rule for ocr_recognize_text: 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 Office Editor. Nothing to install.
ocr_recognize_text 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 ocr_recognize_text 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 ocr_recognize_text. 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.
ocr_recognize_text is provided by the Office Editor MCP server (thewdy/office-editor-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 88 Office Editor tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
88 Office Editor tools catalogued and risk-classified — across an index of 42,500+ MCP servers.