AI agents invoke zahori_validate to trigger actions in Zahori. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool executes an external validation process: it fetches a stream URL, decodes media, and analyzes audio/length. This is not a passive read — it actively triggers network requests and media decoding operations whose effects depend on the provided URL and headers. It fits Execute rather than Read because it performs active processing and external I/O beyond simple data retrieval.
From the tool's definition 'Run the oracle on a resolved stream' and 'decodes, has real audio' — actively fetches and decodes external media content, triggering external network operations and media processing
Attacks that exploit this kind of access
Run the oracle on a resolved stream (url + headers) to check it is real media (decodes, has real audio, plausible length). It is categorised as a Execute tool in the Zahori MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Zahori MCP server in PolicyLayer and add a rule for zahori_validate: 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 Zahori. Nothing to install.
zahori_validate is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the zahori_validate 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 zahori_validate. 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.
zahori_validate is provided by the Zahori MCP server (josesepulvedapino/zahori). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
zahori_validate is one line of Zahori's registry record.
The record carries the whole server: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.
Teams ship this data inside their own products. See what a licence covers →