Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@MCP Tool Auditor Audit the description for my 'search' tool"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Audit any MCP tool description for LLM reliability. Detect smells, scores quality 0-100, and rewrite for better agent accuracy.
Zero external APIs. Pure computation. Built for developers who care about agent reliability.
Research in 2026 shows that 97% of MCP tool descriptions contain at least one "smell" (vague purpose, missing usage guidelines, opaque parameters). When descriptions are weak, AI agents pick the wrong tool or waste steps.
This MCP server lets you (or your agent) audit and fix those descriptions before you publish.
Related MCP server: dissid-audit-mcp
# Install & run with any MCP client
npx -y mcpize connect @princeruhulofficial/tool-auditor --client claudeOr add to your Claude / Cursor / other client config:
{
"mcpServers": {
"tool-auditor": {
"command": "npx",
"args": ["-y", "mcp-tool-auditor"]
}
}
}Tool | What it does |
| Score one tool description (0-100), list smells, give fix suggestions |
| Generate a high-quality rewritten description ready to paste |
| Audit a full list of tools — detect overcrowding + rank quality |
| Return the current best-practice checklist for MCP tool descriptions |
If you are building AI products or selling MCP servers:
Agents that pick the wrong tool = wasted tokens + angry users
High-quality tool descriptions = higher task success rate
This server is free to run (no API bills) and can be called by your own agents during development
Free: 100 audits / day
Pro $9/mo: 2 000 audits / day
npm install
npm run build
npm run test:smokeMIT © Prince Ruhul / Prevalid
Audit a single MCP tool description for LLM reliability. Returns a 0-100 quality score, list of detected smells (vague purpose, missing usage guidelines, opaque params, etc.), and concrete fix suggestions. Use this before publishing any MCP server.
| Name | Required | Description | Default |
|---|---|---|---|
| name | Yes | The exact name of the tool (e.g. 'search_web') | |
| description | Yes | The current human-readable description of the tool | |
| inputSchema | No | The JSON Schema of the tool's input parameters (optional but recommended for deeper analysis) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden. It explicitly discloses the return value (0-100 quality score, list of detected smells, concrete fix suggestions), giving a transparent picture of behavior. It does not mention side effects or permissions, but as a read-only audit tool, the output description is sufficient for most contexts.
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 exactly two sentences: the first states the purpose, the second states the output and usage context. It is front-loaded, concise, and every sentence adds value with no superfluous content.
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?
For a tool with 3 parameters and no output schema, the description explains the output format (score, smells, fixes) and the context (before publishing). It lacks explicit error handling or limitations, but given the simplicity of an audit tool, this is adequate. A slightly higher score would require explicitly stating it is read-only or defining 'LLM reliability.'
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%, meaning all parameters are already documented in the schema. The description does not add parameter-specific details beyond the overall purpose, which implies the 'description' is the main artifact being audited. This meets the baseline for high schema coverage.
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 audits a single MCP tool description for LLM reliability, using the specific verb 'Audit' and specifying the resource ('single MCP tool description'). This distinguishes it from sibling tools like audit_tool_list (likely batch auditing) and rewrite_tool_description (rewriting descriptions).
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 provides a clear usage scenario: 'Use this before publishing any MCP server.' It implies the tool is for pre-publication validation. While it does not explicitly name alternatives or list when-not-to-use conditions, the qualifier 'single' differentiates it from batch tools like audit_tool_list, offering adequate guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Audit an entire list of tools at once. Detects overcrowding (too many tools hurt accuracy), ranks each tool by quality score, and flags the worst offenders. Research shows accuracy drops below 90% when weaker models see more than 10-15 tools.
| Name | Required | Description | Default |
|---|---|---|---|
| tools | Yes | Array of tool objects to audit |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full behavioral burden. It goes beyond a simple statement of action by explaining the logic behind the tool: detecting overcrowding with a specific research-based accuracy threshold and ranking/flags. This provides meaningful context about what the tool does and why, though it does not delve into edge cases or exact output formatting.
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, front-loaded with the core purpose, and each sentence adds value. The second sentence provides supporting evidence for the 'overcrowding' check, making the description both informative and concise without waste.
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?
For a tool with a single parameter and no annotations, the description is reasonably complete: it explains what the tool does, why (research-based threshold), and what outputs to expect (ranking, flags). However, without an output schema, it could be more explicit about the exact structure of the returned data, but the description's mention of ranking and flags covers the essentials.
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?
The schema already describes the only parameter ('tools') as 'Array of tool objects to audit,' which provides 100% coverage. The tool description does not add additional parameter-level meaning beyond what the schema provides, so the baseline score of 3 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 with a specific verb and resource: 'Audit an entire list of tools at once.' It distinguishes itself from sibling 'audit_tool' by emphasizing the batch aspect, and enumerates key capabilities (detect overcrowding, rank by quality score, flag worst offenders).
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 makes it clear this tool is for auditing a list of tools (batch processing), which gives a clear usage context. However, it does not explicitly mention alternatives or exclusion criteria, such as 'for a single tool, use audit_tool instead.' The context is clear but lacks explicit differentiation from sibling tools.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Return the current checklist of MCP tool-description best practices derived from research (2026 papers + real-world agent testing). Use this as a reference when writing new tools.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the burden. It clearly implies a read-only operation ('Return the current checklist') and adds context about the source (research + testing). However, it does not disclose details like response format, potential size, or any limitations, which would enhance transparency for a tool with no annotations.
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, front-loaded with the core purpose, and includes only essential information. Every phrase earns its place, making it highly concise and well-structured.
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 tool is simple (0 params, no output schema), and the description explains what it returns and when to use it. It could add the return format, but the term 'checklist' sufficiently implies a list, making the description complete enough for its low complexity.
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?
The input schema is empty (0 parameters), and the schema description coverage is 100%, so the baseline is 4. The description appropriately avoids mentioning parameters since there are none, and it adds no unnecessary parameter-related details.
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 uses a specific verb ('Return') and clearly identifies the resource ('checklist of MCP tool-description best practices'). It also distinguishes itself from sibling tools like audit_tool and rewrite_tool_description by focusing on a reference checklist rather than action-oriented operations.
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 clearly states when to use the tool: 'Use this as a reference when writing new tools.' It provides a clear context but does not mention exclusions or alternatives, though the sibling tools are distinct in purpose.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Rewrite a weak tool description into a high-quality, LLM-optimized version that follows current best practices (clear purpose, usage guidelines, parameter hints, limitations). Returns the improved description ready to paste into your MCP server.
| Name | Required | Description | Default |
|---|---|---|---|
| name | Yes | Tool name | |
| context | No | Optional extra context about when the tool should be used | |
| description | Yes | Current (weak) description | |
| inputSchema | No | JSON Schema of parameters (helps generate better parameter guidance) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the transparency burden. It discloses the output ('Returns the improved description ready to paste into your MCP server') but doesn't disclose any limitations or side effects, though it's likely a pure text transformation.
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, front-loads the main purpose, and includes no unnecessary information.
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 tool has a simple purpose, 4 parameters with full schema coverage, and the description explains the return value. It could benefit from explicit usage guidance relative to sibling tools, but it is otherwise adequate.
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 coverage is 100%, so each parameter has a description. The tool description mentions best practices that incorporate parameter hints, but doesn't add specific parameter semantics beyond the schema.
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 explicitly states 'Rewrite a weak tool description into a high-quality, LLM-optimized version', which clearly identifies the action and resource. It also distinguishes itself from sibling tools by focusing on rewriting rather than auditing or fetching best practices.
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 implies use when a tool description is weak and needs improvement, by saying 'Rewrite a weak tool description'. It doesn't explicitly mention alternative tools or exclusions, but the context is clear enough.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0audit_toolaudit_tool_listget_best_practicesrewrite_tool_descriptionScored across 4 tools
Audit_tool and audit_tool_list overlap in purpose (auditing descriptions) but are clearly separated by scope (single vs. list). Rewrite and best practices are distinct. The descriptions make the boundaries clear, so confusion is unlikely.
All tool names follow a consistent verb_noun pattern in snake_case (audit_tool, rewrite_tool_description, audit_tool_list, get_best_practices). No mixed casing or stylistic deviations.
Four tools is well-scoped for the server's purpose. Each tool addresses a distinct step in the auditing workflow: reference, single audit, batch audit, and rewrite.
The core workflow is covered: get best practices, audit individual or list, and rewrite weak descriptions. A minor gap is lack of a validation/apply tool after rewriting, but the existing set supports the main use case effectively.
Statically audits MCP tool surfaces for token cost, schema quality, and design issues.
Score MCP tool definitions 0-100. Checks names, schemas, annotations, and uniqueness.
Audits MCP tool definitions for patterns that make models call tools wrong or mis-fill args.
Find, compare, and audit software for AI agents. Scored registry of tools and MCP servers.