Read and extract text from a file. Handles plain text, PDF (via PDFKit), images (via OCR), and documents (.docx, .rtf, .pages via textutil). Text capped at 1MB.
AI agents call files.read to retrieve information from Orchard without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and queries file content without side effects. It performs read-only operations on files (text extraction, OCR) with a 1MB text cap, making it a classic Read category tool with minimal risk if misused by an agent.
From the tool's definition Tool name 'files.read' and description states 'Read and extract text from a file' with support for plain text, PDF, images, and documents. No modification, deletion, or execution capabilities mentioned.
Documented attack patterns abuse exactly the kind of access files.read gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Orchard, and nothing reaches the server without passing your rules. This is the rule we recommend for files.read:
{
"version": "1",
"default": "deny",
"tools": {
"files.read": {}
}
}files.read 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.
Read and extract text from a file. Handles plain text, PDF (via PDFKit), images (via OCR), and documents (.docx, .rtf, .pages via textutil). Text capped at 1MB. It is categorised as a Read tool in the Orchard MCP Server, which means it retrieves data without modifying state.
Register the Orchard MCP server in PolicyLayer and add a rule for files.read: 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 Orchard. Nothing to install.
files.read 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 files.read 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 files.read. 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.
files.read is provided by the Orchard MCP server (l22-io/orchard-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Orchard, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
65 Orchard tools catalogued and risk-classified — across an index of 43,000+ MCP servers.