perception_classify

SERVERMcp Agentcore SOURCEhttps://packagesmcp-perception-production.up.railway.app/mcp

High RISK CLASS

Category Execute

Parameters 42 required

Recommended Rate-limitedsee the rule below

Registry record Grade D, identity unverified Pull the record →

This record as markdown: /tools/com-geiant-mcp-agentcore/perception-classify.md

WHAT IT DOES

What perception_classify does on Mcp Agentcore

AI agents invoke perception_classify to trigger actions in Mcp Agentcore. 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.

Parameter	Type	Required	Description
`task`	string	Yes	Classification task. Currently only "flood" is supported.
`h3_cell`	string	—	Override H3 cell. Defaults to the cell from the original fetch.
`tile_id`	string	Yes	tile_id from perception_fetch_tile result (must be in session cache).
`write_to_spatial_memory`	boolean	—	Write perception chain to geiant_geometry_state. Default: true.

Parameters from the server's own tool schema.

RISK

Why perception_classify is rated High

This tool executes an external machine learning inference pipeline on satellite imagery data and modifies system state by writing results to Spatial Memory and the agent audit trail. While the core action is a read of image data and classification, the invocation of a remote compute service (RunPod endpoint) to execute a specialized model constitutes Execute-category behavior.

From the tool's definition 'Run Prithvi-EO-2.0-300M-TL-Sen1Floods11 flood classification' and 'Sends the 6-band chip to a RunPod endpoint' indicate execution of an external ML model.

Attacks that exploit this kind of access

RECOMMENDED RULE

The rule that runs perception_classify safely

PolicyLayer is an MCP gateway: it sits between your AI agents and Mcp Agentcore, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For perception_classify, this is the rule to start with:

perception_classify Rate-limited

perception_classify 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.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "perception_classify": {
      "limits": [
        {
          "counter": "perception_classify_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

RATE-LIMIT THIS TOOL → Instant setup, no code required.

The button opens the PolicyLayer dashboard: create your workspace, connect Mcp Agentcore, apply this rule, and every perception_classify call is checked against it from then on.

FAQ

Questions about perception_classify

What does the perception_classify tool do? +

Run Prithvi-EO-2.0-300M-TL-Sen1Floods11 flood classification on a Sentinel-2 tile previously fetched by perception_fetch_tile. Sends the 6-band chip to a RunPod endpoint and returns: dominant_class, flood_pixel_pct, confidence, class_counts, and the full perception_chain. The perception chain is written to Spatial Memory and a signed audit breadcrumb is dropped to the agent trail. It is categorised as a Execute tool in the Mcp Agentcore MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

What parameters does perception_classify accept? +

perception_classify accepts 4 parameters: task, h3_cell, tile_id, write_to_spatial_memory. Required: task, tile_id. The full parameter table on this page comes from the server's own tool schema.

How do I enforce a policy on perception_classify? +

Register the Mcp Agentcore MCP server in PolicyLayer and add a rule for perception_classify: 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 Mcp Agentcore. Nothing to install.

What risk level is perception_classify? +

perception_classify is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit perception_classify? +

Yes. Add a rate_limit block to the perception_classify 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.

How do I block perception_classify completely? +

Set action: deny in the PolicyLayer policy for perception_classify. 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.

What MCP server provides perception_classify? +

perception_classify is provided by the Mcp Agentcore MCP server (https://packagesmcp-perception-production.up.railway.app/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

KEEP EXPLORING

More on Mcp Agentcore, and thousands of servers like it.

This server

All 8 Mcp Agentcore tools → High-risk tools on this server →

Across the catalogue

All high-risk MCP tools → The MCP Attack Database →

Guides

Govern code and CI agents →Agent sandbox →Least privilege for MCP →