ikeytz - Schlüsseldienst Ludwigsburg

Get Llms Jsonld

get_llms_jsonld

Read-only

WHAT: Fetch /llms-jsonld.txt — map every marketing HTML URL to JSON-LD relatedLink (ai-pages fulltext) + mirrorId + optional geo_api/geo_grenze/geo_punkt (maps Embed-API). Format: pageUrl TAB relatedLink TAB mirrorId TAB geo_api TAB geo_grenze TAB geo_punkt. Ort = one URL each; PLZ = pipe-separated URLs per Ort of that PLZ (L-30). Same engine as get_sitemap_txt (offset/limit/nextOffset/# MCP-META). Not the page body (use get_ai_page). Not get_serp_snippet. Geo columns empty for Hub/Legal/non-ort.

Input Schema

Name	Required	Description	Default
`limit`	No	Maximum number of lines to return after offset. Omit or null = if remaining payload ≤~30KB return all remaining lines; if larger, return first ~30KB window + nextOffset (O9). Explicit limit keeps 1..10000 line window (MCP caps unchanged — not forced to 30KB). Meta also as trailing # MCP-META lines in summary.
`offset`	No	0-based line index into the fetched UTF-8 file (split on \n). 0 = first line. Omit = 0. Combined with limit = a sliding window. Use nextOffset from the previous result to page. Does not count bytes; one line can be a long URL.

Output Schema

Name	Required	Description
`ok`	Yes	true = use summary, canonicalUrl, and extra keys. false = do not invent data; read error + hint and retry with valid args or open hint.
`hint`	No	Present when ok=false. Absolute URL the agent should open or pass to get_ai_page / get_discovery.
`path`	No	Public path on www, e.g. /llms.txt or /.well-known/ai.txt
`tool`	Yes	Echo of the tool name that produced this object (e.g. get_service_area).
`error`	No	Present when ok=false. Codes include: unknown_tool, unknown_place, unknown_auswahl, unknown_ratgeber, unknown_doc, unknown_discovery, faq_not_found, wissen_not_found, serp_not_found, q_required, bad_datetime, bad_when_key, plus HTTP errors from fetch.
`limit`	No	Applied line limit, or null if the caller omitted limit.
`which`	No	Resolved discovery id after alias fold (llms, sitemap-txt, ard, ai-catalog, …).
`locale`	No	Locale actually used for URLs (args.locale or de).
`offset`	No	Applied 0-based line offset.
`related`	No	Always includes https://www.ikeytz.com/llms.txt. Geo tools also include https://maps.ikeytz.com/llms.txt.
`summary`	No	Primary text for the model. For discovery tools this is the file window (possibly thousands of characters). For list tools a compact one-liner. Quote with attribution.
`totalUrls`	No	http(s) lines in the loaded file (full count when public/ read succeeds).
`truncated`	No	true if load hit a byte cap OR offset+limit left more lines. Also mirrored in trailing # MCP-META lines in summary.
`nextOffset`	No	offset + returnedLines when more lines remain in the loaded file; else null. Pass as the next offset. Also in # MCP-META.
`totalBytes`	No	Uncompressed UTF-8 byte length of the discovery file (decoded body). Content-Length is used only for identity responses when it is ≥ body size (full file if the load cap truncated). Gzip wire Content-Length is ignored. Also in # MCP-META.
`totalChars`	No	Character length of the downloaded body before applying discoveryFetchMax (may equal totalBytes for ASCII).
`totalLines`	No	Line count of the loaded file (full public/ file when available).
`attribution`	Yes	Required citation: 'Quelle: Schlüsseldienst Ludwigsburg ikeytz (ikeytz.com) · ai-train=no · https://www.ikeytz.com/.well-known/ai.txt'
`canonicalUrl`	No	Best URL to show the user: https://www.ikeytz.com… page, or tel:+49…, mailto:…, https://wa.me/…. Prefer this over constructing URLs.
`returnedUrls`	No	Lines in this window that start with http(s)://
`bytesReturned`	No	UTF-8 byte length of the returned body window (before # MCP-META).
`returnedLines`	No	How many lines are in summary/body this call (before # MCP-META).
`fetchTruncated`	No	true if discoveryFetchMax cut the load (line-boundary safe). Large discovery files (sitemap-xml ~12MB, llms-keywords ~55MB, llms-index ~2MB) must load full so offset/limit can reach EOF.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 10, 2026

changedOutput schema / properties / totalBytes / description

Previous value: -"Full origin file size when Content-Length is present (e.g. llms-keywords ~50MB); else character length of the downloaded body before the load cap. Also in # MCP-META."New value: +"Uncompressed UTF-8 byte length of the discovery file (decoded body). Content-Length is used only for identity responses when it is ≥ body size (full file if the load cap truncated). Gzip wire Content-Length is ignored. Also in # MCP-META."

Changed1 schema field changedOct 10, 2026

changedInput schema / properties / limit / description

Previous value: -"Maximum number of lines to return after offset. Omit or null = return every remaining line in the loaded file. Clamped to 1..10000. For sitemap.txt (~9076 http-URLs, Stand 2026-09-22) omit limit to get the full list (or page with offset/limit + nextOffset). For llms-keywords.txt prefer a window; the file is huge. Meta also as trailing # MCP-META lines in summary."New value: +"Maximum number of lines to return after offset. Omit or null = if remaining payload ≤~30KB return all remaining lines; if larger, return first ~30KB window + nextOffset (O9). Explicit limit keeps 1..10000 line window (MCP caps unchanged — not forced to 30KB). Meta also as trailing # MCP-META lines in summary."

Changed1 schema field changedSep 25, 2026

changedOutput schema / properties / fetchTruncated / description

Previous value: -"true if discoveryFetchMax cut the load (line-boundary safe). sitemap-xml (~12MB) and llms-index (~2MB) must load full; keywords stay capped — use totalBytes."New value: +"true if discoveryFetchMax cut the load (line-boundary safe). Large discovery files (sitemap-xml ~12MB, llms-keywords ~55MB, llms-index ~2MB) must load full so offset/limit can reach EOF."

Changed1 schema field changedSep 25, 2026

changedOutput schema / properties / fetchTruncated / description

Previous value: -"true if discoveryFetchMax cut the load (line-boundary safe). sitemap-xml must load full (~12MB cap); keywords stay capped — use totalBytes."New value: +"true if discoveryFetchMax cut the load (line-boundary safe). sitemap-xml (~12MB) and llms-index (~2MB) must load full; keywords stay capped — use totalBytes."

Changed3 schema fields changedSep 24, 2026

changedOutput schema / properties / fetchTruncated / description

Previous value: -"true if a byte cap cut the load (line-boundary safe). Rare for sitemap.txt (reads public/)."New value: +"true if discoveryFetchMax cut the load (line-boundary safe). sitemap-xml must load full (~12MB cap); keywords stay capped — use totalBytes."

addedOutput schema / properties / totalBytes

Added value: +{
+  "description": "Full origin file size when Content-Length is present (e.g. llms-keywords ~50MB); else character length of the downloaded body before the load cap. Also in # MCP-META.",
+  "type": "number"
+}

addedOutput schema / properties / totalChars

Added value: +{
+  "description": "Character length of the downloaded body before applying discoveryFetchMax (may equal totalBytes for ASCII).",
+  "type": "number"
+}

AddedSep 24, 2026

TDQS

A4.1/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already cover readOnlyHint=true, and the description adds substantial context beyond that: the payout format, the shared engine with get_sitemap_txt (offset/limit/nextOffset/# MCP-META), and that geo columns are empty for Hub/Legal/non-ort. This is exactly the extra behavioral context annotations do not carry.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is dense and front-loaded with WHAT and the format, then exclusions. It is long, but nearly every clause encodes a distinct fact; the one weak spot is the 'Ort/PLZ' glossary shorthand which assumes site-specific vocabulary an agent may not have.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a read-only file-fetch tool with a full output schema, the description covers format, engine behavior, paging meta, and domain exceptions. The only gap is that the Ort/PLZ distinction is explained in terse German shorthand rather than plain terms, which could leave a non-domain-aware agent guessing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% and both parameters have rich descriptions including the ~30KB window rule and offset/nextOffset paging. The description only restates the paging engine by reference, adding nothing beyond the schema, so the baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description names a specific artifact (/llms-jsonld.txt), enumerates exactly what it maps (marketing HTML URL → relatedLink + mirrorId + optional geo columns), and gives the tab-separated column layout. It explicitly distinguishes from get_ai_page and get_serp_snippet, so an agent can tell it apart from siblings without opening a schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It states when-not to use it ('Not the page body (use get_ai_page). Not get_serp_snippet') and explains the Ort vs PLZ line semantics (one URL each vs pipe-separated per Ort). It does not say which sibling to use when the goal is the JSON-LD itself versus other llms_* variants, so a 5 is not warranted.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All ikeytz - Schlüsseldienst Ludwigsburg tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?