Create, validate and audit llms.txt, incl. the Lighthouse Agentic Browsing check.
Scored across 4 tools
Each tool has a distinct role: generate creates a draft, validate scores arbitrary content against llmstxt.org, check_lighthouse applies exact Lighthouse rules, and check_published fetches a live URL. The only mild overlap is that validate also reports Lighthouse status, but the descriptions clarify the different focus.
Names are all snake_case verb-first phrases such as generate_llms_txt and check_published_llms_txt. check_lighthouse_agentic_browsing breaks the llms_txt-noun pattern slightly and is awkward, but the verb-first convention is otherwise consistent.
With four tools, the server is tightly scoped to the llms.txt workflow: creation, validation, Lighthouse auditing, and live publishing checks. Each tool addresses a clear need without bloat.
The set covers the full practical lifecycle: generate a draft, validate against the standard, check Lighthouse compliance, and verify the published file. No obvious dead ends or missing core operations.
Prüft einen llms.txt-Inhalt exakt gegen die drei Regeln des Lighthouse-Audits llms-txt (Lighthouse 13.3): H1 vorhanden, mindestens ein Markdown-Link, mehr als 50 Zeichen. Benennt bei Fehlschlag die Originalmeldung von Lighthouse.
| Name | Required | Description | Default |
|---|---|---|---|
| content | Yes | Der Inhalt der llms.txt. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The readOnlyHnt annotation already signals a safe read operation, and the description adds real behavioral detail: exactly which three rules are checked and that failures report the original Lighthouse message. This goes beyond the annotations and does not contradict them.
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 a single dense sentence that front-loads the core action, then lists the exact rules and failure behavior. There is no filler or redundancy; every clause contributes necessary 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?
For a simple tool with one fully documented string parameter, the description is mostly complete: it states the audit rules and what happens on failure. The only gap is the unspecified success return format, but since no output schema exists and the failure behavior is described, an agent can still select and invoke the tool correctly.
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 already describes the single 'content' parameter fully, including its maxLength, and schema description coverage is 100%. The tool description adds no new parameter-level detail, 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 names a specific verb ('prüft'), the exact resource ('llms.txt-Inhalt'), and the precise audit rules (H1, one Markdown link, >50 characters) from a named audit version (Lighthouse 13.3). This clearly distinguishes it from sibling tools like validate_llms_txt or check_published_llms_txt.
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 does not explicitly name alternative tools or exclusions, but the context is clear: use this when you need to check an llms.txt content against the Lighthouse llms-txt audit rules. The usage context is implied strongly enough for an agent to select it correctly, though explicit sibling routing would be better.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Ruft die live ausgelieferte /llms.txt einer Domain ab und prüft HTTP-Status, Content-Type und Inhalt. Nutzen Sie dies, um zu klären, warum PageSpeed Insights eine llms.txt bemängelt.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | Domain oder beliebige URL der Website, z. B. "beispiel.de". |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations provide readOnlyHint and openWorldHint, and the description adds meaningful behavioral detail: it fetches the live /llms.txt and evaluates HTTP status, content-type, and content. This goes beyond the annotations by explaining exactly what the tool inspects, though it doesn't disclose output format or response behavior.
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?
Two concise, information-dense sentences. The first sentence states the core operation and checked aspects, and the second provides the practical use case. No filler or redundant repetition.
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 with one param and clear behavior, but there is no output schema and the description doesn't state what the tool returns. An agent knows what it checks and why to use it, but not what result shape to expect, which is a notable gap for a diagnostic tool.
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 parameter url is already documented as 'Domain oder beliebige URL der Website'. The description adds little beyond the schema; it implies the URL is the target domain, but that matches the schema. Baseline 3 is appropriate since the schema carries the parameter semantics.
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 and resource: retreives the live-delivered /llms.txt of a domain and checks HTTP status, content-type, and content. It distinguishes itself from siblings by emphasizing 'live ausgelieferte' and the PageSpeed Insights use case, which clearly separates it from generate_llms_txt and validate_llms_txt.
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 gives an explicit trigger for use: 'Nutzen Sie dies, um zu klären, warum PageSpeed Insights eine llms.txt bemängelt.' This is clear context for when the tool should be invoked. However, it doesn't mention alternatives or when not to use it, so it falls short of a full 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Crawlt bis zu 15 öffentliche HTML-Seiten einer Domain und erzeugt daraus einen fertigen llms.txt-Entwurf. Deterministisch: Es wird kein Sprachmodell aufgerufen, Beschreibungen stammen ausschließlich aus den Metadaten oder sichtbaren Texten der Quellseiten. Liefert zusätzlich Validierung, Lighthouse-Status und die Liste der ausgeschlossenen URLs mit Begründung. Die Eingabe darf eine blanke Domain sein.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | Website, z. B. "beispiel.de" oder "https://www.beispiel.de". | |
| maxPages | No | Obergrenze der Seiten, Standard 15. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond readOnlyHint/openWorldHint, the description discloses the deterministic no-LLM behavior, the 15-page crawl cap, the metadata/visible-text sourcing of descriptions, and the auxiliary outputs (validation, Lighthouse status, excluded URLs with reasons). This materially shapes an agent's expectations about cost and output quality.
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 sentences with the main action front-loaded in the first. Each subsequent sentence adds a distinct fact: determinism, delivered outputs, and input flexibility. No filler or repetition of schema content beyond minimal reinforcement.
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 2-parameter tool with no output schema, the description covers input, process, determinism, and the main return components (validation, Lighthouse status, excluded-URL list). A minor gap is the absence of failure/edge-case behavior for unreachable domains or crawl exhaustion.
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 baseline is 3; the schema already documents accepted url formats and the maxPages range (3–25, default 15). The description adds only marginal reinforcement by echoing the 15-page cap and clarifying that a bare domain is acceptable input.
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 first sentence names a specific verb+resource: it crawls up to 15 public HTML pages of a domain and produces a finished llms.txt draft. The determinism qualifier ('kein Sprachmodell aufgerufen') further distinguishes it from a generic LLM-based generator and from the sibling validation/check tools.
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 establishes clear selection context: use this to generate an llms.txt draft from a domain, and the determinism note ('Beschreibungen stammen ausschließlich aus den Metadaten oder sichtbaren Texten') implicitly tells agents not to use it when LLM-generated descriptions are desired. However, siblings like validate_llms_txt or check_published_llms_txt are never named, so the when-not-to-use guidance is implied rather than explicit.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Prüft den Inhalt einer llms.txt gegen den Formatvorschlag von llmstxt.org: H1, Zusammenfassung, H2-Abschnitte, absolute Links, Duplikate, Links auf fremde Domains, Dateigröße und Aktualisierungsdatum. Liefert 0–100 Punkte sowie Befunde mit Schweregrad und Zeilennummer, dazu den Lighthouse-Status.
| Name | Required | Description | Default |
|---|---|---|---|
| content | Yes | Der vollständige Inhalt der llms.txt. | |
| expectedHostname | No | Hostname der eigenen Website; aktiviert die Warnung bei Links auf fremde Domains. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description goes beyond the readOnlyHint annotation by disclosing concrete evaluation criteria and the output shape: a 0–100 score, findings with severity and line numbers, and Lighthouse status. It is strong but leaves the meaning of 'Lighthouse-Status' somewhat unexplained.
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 one dense, efficient sentence that front-loads the purpose, lists the validation scope, and states the output. Every clause earns its place and there is no filler.
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?
With no output schema, the description usefully explains return values and validation criteria. It is complete enough to invoke the tool, but it could have clarified the role of expectedHostname relative to external-domain checks and explained the Lighthouse status more fully.
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 parameters are already well documented. The description adds context about what is validated but does not add new meaning to the parameters themselves, which matches the baseline for full 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 names a clear verb and resource: 'Prüft den Inhalt einer llms.txt gegen den Formatvorschlag von llmstxt.org' and enumerates the exact checks performed. It is specific enough to distinguish this from publishing or generating a llms.txt, even though it does not name the sibling tools explicitly.
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 that the tool is for validating llms.txt content, but it gives no explicit guidance about when to prefer it over siblings like check_published_llms_txt or generate_llms_txt. No exclusions, prerequisites, or alternative routing are mentioned.
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.
check_lighthouse_agentic_browsingcheck_published_llms_txtgenerate_llms_txtvalidate_llms_txtAudit a page for search and AI answer engines; generate robots.txt, sitemap, head, llms.txt.
Checks llms.txt, AI crawler access in robots.txt, and sitemap - with a 0-100 AI readiness score.
Check website AI-readiness: Schema.org, llms.txt, E-E-A-T, robots.txt. Works in Cursor & Claude.
Generate 18 AI readiness files (llms.txt, ai.txt, RAG indexes, schema) for any website.
Add one secure layer between your agents and this server.
