vision.track
Associate detections to tracks for one frame (IoU + appearance + Kalman).
This record as markdown: /tools/visionmcp/vision.track.md
What vision.track does on Visionmcp
AI agents invoke vision.track to trigger actions in Visionmcp. 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.
Why vision.track is rated High
This tool runs a tracking algorithm (IoU matching, appearance features, Kalman filter) to associate detections with tracks, which constitutes executing a computation/processing pipeline. It modifies in-memory tracking state rather than simply reading data, but is not a persistent write to a database or destructive action. Severity is medium because misuse could corrupt tracking state in an active vision pipeline.
From the tool's definition Associate detections to tracks for one frame (IoU + appearance + Kalman)
Attacks that exploit this kind of access
The rule that runs vision.track safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Visionmcp, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For vision.track, this is the rule to start with:
vision.track stays usable, but rate-capped: a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Visionmcp, apply this rule, and every vision.track call is checked against it from then on.
Questions about vision.track
Associate detections to tracks for one frame (IoU + appearance + Kalman). It is categorised as a Execute tool in the Visionmcp MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Vision MCP server in PolicyLayer and add a rule for vision.track: 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 Visionmcp. Nothing to install.
vision.track 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 vision.track 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 vision.track. 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.
vision.track is provided by the Vision MCP server (joshuahickscorp/visionmcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Vision, and thousands of servers like it.
Across the catalogue