AI agents call verify_file to retrieve information from Unfakable without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves or queries verification data (whether a proof exists for a file hash) and performs read-only cryptographic operations. The explicit statement that 'the file is never uploaded — only its hash travels' confirms no side effects occur. The operation is purely informational and non-destructive, making it a Read category tool with low severity even in adversarial scenarios.
From the tool's definition The tool 'Hash a local file with SHA-256 and check whether Unfakable has a Bitcoin-anchored proof for that fingerprint.
Attacks that exploit this kind of access
Hash a local file with SHA-256 and check whether Unfakable has a Bitcoin-anchored proof for that fingerprint. The file is never uploaded — only its hash travels. It is categorised as a Read tool in the Unfakable MCP Server, which means it retrieves data without modifying state.
Register the Unfakable MCP server in PolicyLayer and add a rule for verify_file: 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 Unfakable. Nothing to install.
verify_file 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 verify_file 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 verify_file. 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.
verify_file is provided by the Unfakable MCP server (unfakable/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 →