Check whether an MCP server's tools have changed since they were first recorded, and whether any tool description contains prompt-injection or data-exfiltration patterns. Call this BEFORE trusting a newly added or recently updated MCP server. Returns a verdict of clean or REVIEW_BEFORE_USE.
| Name | Required | Description | Default |
|---|---|---|---|
| url | No | The MCP server's HTTP endpoint to probe live. | |
| server | No | Registry name of the server, if the URL is unknown. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It discloses that the tool performs a check and returns a verdict, and implies probing a live server via the 'url' parameter. However, it does not explicitly state that the operation is read-only, does not modify anything, or disclose any safety considerations (e.g., network access requirements, potential side effects). This is a moderate level of transparency.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences long, front-loaded with the core purpose, followed by usage guidance and the return value. Every sentence contributes meaningful information with no redundancy or filler. It is a model of concise, well-structured documentation.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The description explains what the tool does, when to use it, and what it returns (a verdict). However, it lacks details about how to choose between the 'url' and 'server' parameters, whether both can be provided, and what happens if neither is given. Given there is no output schema, the return value is described adequately, but the parameter selection logic is a gap in completeness.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% – both 'url' and 'server' have clear descriptions in the schema. The tool description itself does not add any parameter semantics beyond that. Baseline for high coverage is 3, and there is no additional explanation of how the parameters interact (e.g., whether they are mutually exclusive or preferred), so this score is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: checking if an MCP server's tools have changed since first recorded and scanning for prompt-injection or data-exfiltration patterns. It uses specific verbs ('check', 'returns a verdict') and distinguishes itself from siblings like get_drift_feed and get_findings, which are about retrieving feeds/findings rather than assessing a server's trustworthiness.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly instructs when to use this tool: 'Call this BEFORE trusting a newly added or recently updated MCP server.' This is a clear usage context. However, it does not mention any alternatives or explicitly state when not to use it, so it doesn't fully meet the 'explicit when/when-not/alternatives' bar for a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Each tool has a distinct purpose: check_mcp_server targets a single server for both changes and injection patterns, get_drift_feed scans the ecosystem for recently changed tools, and get_findings lists tools with detected patterns. No two tools are likely to be confused.
All tool names follow a consistent verb_noun pattern using snake_case: check_mcp_server, get_drift_feed, get_findings. The verbs 'check' and 'get' are both action-oriented and appropriate, maintaining predictable naming.
Three tools is a compact but reasonable set for a monitoring server focused on MCP ecosystem changes and security. While slightly lean, each tool covers a core concern without redundancy.
The surface covers the primary monitoring actions: checking a specific server, browsing ecosystem drift, and retrieving findings. A minor gap is the lack of a detailed drill-down for why a specific server is flagged beyond the initial verdict, but workflows are still functional.