Check untrusted text (user messages, retrieved pages, tool outputs) for prompt injection before it reaches the LLM. Returns safe or not, plus the threat type. Free tier of 10,000 checks a month.
Scored across 2 tools
The two tools are functionally distinct (single input vs. batch), and descriptions make intended usage clear. However, the names differ only by a trailing 's', which is a real misselection risk for an agent that pattern-matches on names rather than reading the descriptions.
Both tools use consistent snake_case verb_noun form with a shared 'validate_prompt' stem, so the convention is predictable. The near-identical singular/plural pair is a minor readability hazard but follows an established API idiom.
Two tools is thin for the surface, though the server's scope (a focused prompt-injection classifier) justifies a minimal API. The single/bulk split is the only natural axis of variation, so little is missing.
Single and batch validation cover the core lifecycle for this domain, including bulk pipelines and RAG indexing. Gaps are minor: no way to inspect threat categories/taxonomy, configure sensitivity thresholds, or check server health.
Check a single piece of untrusted text (a user message, a retrieved RAG document, tool output, etc.) for prompt injection, jailbreaks, instruction-override, data-exfiltration, and code injection BEFORE passing it to an LLM. Returns a safe/unsafe verdict with confidence, threat category, and reasoning.
| Name | Required | Description | Default |
|---|---|---|---|
| prompt | Yes | The untrusted text to validate. | |
| sensitivity | No | Optional. Omit to use the account default. 'no_judgement' blocks only clear attacks, for apps whose users tell the AI exactly what to write. 'balanced' is an older name for it. 'strict' and 'lenient' are accepted for existing callers. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Lacking annotations, the description provides rich behavioral context by detailing exactly what is checked (e.g., data-exfiltration, code injection) and what the response structure looks like (verdict, confidence, category, reasoning).
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 tool definition is highly concise, delivering maximum information—scope, purpose, and result expectations—in a single, dense, and well-structured paragraph.
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?
Even without an output schema, the description fully outlines the functionality, listing the specific types of threats addressed and the detailed, categorized nature of the returned verdict.
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 defines the sensitivity parameter, but the description provides crucial semantic context by explaining the operational differences between 'no_judgement,' 'balanced,' and 'strict' modes.
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 validates untrusted text against multiple, specific security threats (e.g., injection, jailbreaks). It clearly defines the action, though it only vaguely differentiates itself from the plural sibling, validate_prompts.
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?
It provides a clear operational guideline by stating the validation should happen 'BEFORE passing it to an LLM.' However, it fails to offer any guidance on why one would choose this singular tool over the plural sibling, validate_prompts.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Validate many pieces of untrusted text in one call (processed in parallel server-side). Use for bulk pre-screening, indexing pipelines, or checking a batch of retrieved documents. Returns one verdict per input, in order.
| Name | Required | Description | Default |
|---|---|---|---|
| prompts | Yes | Array of untrusted texts to validate (max 100). | |
| sensitivity | No | Optional. Omit to use the account default. 'no_judgement' blocks only clear attacks, for apps whose users tell the AI exactly what to write. 'balanced' is an older name for it. 'strict' and 'lenient' are accepted for existing callers. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full behavioral burden, and it usefully discloses that processing happens 'in parallel server-side' and that results are returned 'one verdict per input, in order.' It still omits operational details such as failure behavior, rate limits, or whether inputs are retained, so transparency is adequate but incomplete.
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?
Three tight sentences, front-loaded with the core action and scope, followed by use cases and return behavior. Every sentence contributes and nothing is wasted.
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 simple two-parameter batch validation tool with no output schema, the description covers purpose, use cases, parallel processing, and ordering. It stops short of describing the verdict format or failure semantics, but it is nearly complete for selection and invocation.
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%, so the input schema already documents both the prompts array and the sensitivity enum thoroughly. The description does not add parameter-level detail beyond the schema, which is the expected baseline when schema coverage is high.
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 states a specific verb and resource ('Validate many pieces of untrusted text') and makes the batch scope explicit with 'in one call' and 'bulk', which clearly distinguishes it from the singular sibling validate_prompt. An agent can immediately tell this is the multi-item variant.
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?
It gives concrete usage contexts: 'bulk pre-screening, indexing pipelines, or checking a batch of retrieved documents.' It does not explicitly say when not to use it or name validate_prompt as the alternative for single inputs, but the batch-oriented guidance is clear.
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.
validate_promptvalidate_prompts
Add one secure layer between your agents and this server.