Search 13,000+ remote/hybrid jobs in Germany/EU, open a posting, get market stats. Free, no auth.
Scored across 3 tools
The three tools have clearly distinct purposes: search_jobs finds postings, get_job fetches one posting by id, and get_job_statistics returns aggregate board metrics. There is no meaningful overlap, and the descriptions make the boundaries obvious.
All tool names use consistent snake_case with a leading action verb: get_job, get_job_statistics, search_jobs. The pattern is predictable and easy for an agent to follow.
Three tools are well-scoped for a read-only remote-jobs board: search, detail retrieval, and aggregate statistics. Each tool earns its place without redundancy.
The surface covers the core read-only workflow: search listings, inspect a single posting with provenance, and view board statistics. Minor gaps exist around auxiliary metadata such as standalone company/source lookups, but agents can work around them via filters and search.
Fetch ONE nomado24 job posting by id, with the complete v2 payload: per-field provenance (dataOrigin: who asserted each field, by which method, when), salaryOrigin, firstSeenAt, lastVerifiedAt with its verification method, and updatedAt (the last genuine content change). Returns an error result for an id that is not a live posting; expired and withdrawn postings are not served. Read-only, free access. No blanket content licence; retain attribution and respect third-party rights (see structuredContent.attribution).
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | Job id as returned by search_jobs ('job_<slug>'). A bare slug is also accepted |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnly/idempotent/non-destructive, and the description adds substantial context beyond them: error behavior for non-live ids, exclusion of expired and withdrawn postings, free access, and an explicit licensing/attribution obligation pointing at structuredContent.attribution. That is exactly the kind of operational and legal context annotations cannot express.
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 dense sentences, front-loaded with the core action and record scope before the payload detail. The provenance-field enumeration is long but each item is genuinely informative; a small amount of tightening is possible but 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?
With no output schema, the description carries the return-value burden and does so by naming the key payload fields (dataOrigin, salaryOrigin, firstSeenAt, lastVerifiedAt, updatedAt) plus error and licensing behavior. An agent has everything needed to call it and interpret the result.
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% and the single parameter is fully documented there, including the 'job_<slug>' format and bare-slug acceptance. The description adds only "by id" and does not repeat or enrich the format rules, so the baseline 3 applies.
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?
States a specific verb and resource ("Fetch ONE nomado24 job posting by id") and explicitly scopes to a single record, which cleanly separates it from the sibling search_jobs list operation. The payload contents are enumerated, so an agent knows exactly what it gets back.
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?
Clear usage context: it takes an id and returns an error result for ids that are not live postings, with expired and withdrawn postings explicitly not served. It never names search_jobs as the way to obtain an id in the description itself, but the single-record framing plus the schema's "Job id as returned by search_jobs" makes the routing unambiguous.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get aggregate statistics (active jobs, new-this-week/month, top sources, salary ranges) for nomado24's remote-jobs board. No input required.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover the safety profile (readOnlyHint=true, idempotentHint=true, destructiveHint=false), lowering the bar. The description adds useful context by listing the statistics categories returned, but it does not disclose details like data freshness, possible overlap between new-this-week and new-this-month counts, or the response shape.
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?
One front-loaded sentence: the verb and resource come first, followed by a compact parenthetical list of statistics types and a short input note. Every word earns its place and nothing repeats the title or the schema.
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 0-parameter, read-only statistics endpoint with no output schema, the description adequately covers what the tool does, what data it returns, and that no input is needed. The only gap is the absence of explicit routing guidance between this tool and search_jobs, though the aggregate-vs-search contrast is reasonably evident.
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?
With zero parameters and 100% schema coverage, there is nothing for the schema to document. The description reinforces this with 'No input required,' which is exactly the right semantic signal for a no-parameter tool, justifying the baseline-4 score.
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+resource ('Get aggregate statistics... for nomado24's remote-jobs board') and enumerates exactly what is included (active jobs, new-this-week/month, top sources, salary ranges). The aggregate-stats framing clearly separates it from get_job (single job) and search_jobs (search), though it does so implicitly rather than by naming the siblings.
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 only usage guidance is 'No input required,' which correctly tells the agent that no parameters are needed. However, there is no explicit statement of when to choose board-level statistics over search_jobs or get_job; the agent must infer the intent solely from the word 'aggregate.'
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search nomado24's remote/hybrid job board. Read-only, free access. No blanket content licence; retain attribution and respect third-party rights (see structuredContent.attribution). Answers over one of two contracts, chosen by the arguments you send. Send only q/language/page/per_page and you get the legacy v1 result shape with page-based paging. Send any structured filter (country, applicant_region, work_arrangement, employment_type, seniority, skills, salary_min/max, published_after, verified_after, source, company, sort, include_total) or a cursor and you get the v2 contract: richer job objects, per-posting provenance (salaryOrigin, firstSeenAt, lastVerifiedAt, verification.method, updatedAt), a corpus changeWatermark, a snapshotId and keyset paging via pagination.nextCursor. The two are mutually exclusive: page plus a structured filter is refused rather than silently resolved. Structured results omit per-field dataOrigin; call get_job for provenance. Copied posting text and external application URLs are withheld by default unless separately documented redistribution permission exists. A filter nomado24 cannot answer honestly is refused with the reason, never answered with the unfiltered corpus. Structured pagination.total is currently null (totalRelation 'not_computed') for every query, because a current-text on-site-only serving gate cannot be counted accurately by the database. A null total is deliberate, not an error.
| Name | Required | Description | Default |
|---|---|---|---|
| q | No | Free-text search over job title/company/tags | |
| page | No | 1-based page number, default 1 | |
| sort | No | One of published_desc, verified_desc, salary_asc, salary_desc. Default published_desc | |
| cursor | No | Continuation cursor from pagination.nextCursor. Send it ALONE (with per_page at most), never alongside filters: the cursor already carries the filter set it traverses | |
| skills | No | Skill slugs, lowercase kebab-case (react, node-js) | |
| source | No | Feed/source ids, lowercase kebab-case | |
| company | No | Company name, exact match, case-insensitive. Aggregated postings carry no verified employer identity | |
| country | No | Where the ROLE sits, ISO 3166-1 alpha-2 uppercase (DE, US) | |
| language | No | Filter by job content language | |
| per_page | No | Results per page, 1-25, default 20 | |
| seniority | No | Seniority. 'unknown' means the field is absent in the response, not a stored value | |
| salary_max | No | Upper bound in normalized ANNUAL EUR (gross) | |
| salary_min | No | Lower bound in normalized ANNUAL EUR (gross). Matches jobs whose salary interval overlaps it | |
| skills_match | No | How to combine skills. Requires skills | |
| include_total | No | Requested total mode (none, estimated, exact). Totals are currently withheld for every structured search: a current-text on-site-only serving gate is not visible to the database count, even without a named filter. The response reports total: null and totalRelation: 'not_computed' for all three modes. | |
| verified_after | No | RFC 3339 date-time. Compares against the SAME value the response reports as lastVerifiedAt | |
| employment_type | No | Employment type. 'unknown' means the field is absent in the response, not a stored value | |
| published_after | No | RFC 3339 date-time, e.g. 2026-08-01T00:00:00Z | |
| salary_currency | No | ISO 4217. Only EUR: every salary is normalized to EUR at ingest and no conversion is performed | |
| applicant_region | No | Where a CANDIDATE may be: WORLDWIDE, EU or an ISO 3166-1 alpha-2 code. EEA/DACH/EMEA are refused, nomado24 does not classify them | |
| work_arrangement | No | Work arrangement. The board serves remote and hybrid postings only, so these are the only two values |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond the readOnly and idempotent annotations, the description discloses free access, attribution/licence obligations, withheld redistribution content, refusal behavior for unanswerable filters, and the deliberate null total. These are substantial behavioral traits an agent would not know from annotations or schema alone.
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 dense but front-loaded with purpose and contract selection, then layers on provenance, withholding, refusal, and pagination caveats. Every sentence carries operational meaning for a 21-parameter search tool, so the length is justified rather than wasteful.
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 and a large parameter surface, the description supplies the missing context: two response contracts, provenance fields, cursor behavior, total semantics, refusal behavior, and attribution obligations. An agent has enough information to choose arguments and interpret the response contract 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?
Although schema description coverage is already 100%, the description adds contract-level meaning that the schema does not: q/language/page/per_page alone yields v1, any structured filter or cursor yields v2, and page plus filter is refused. It also clarifies that structured results omit per-field dataOrigin and that total is intentionally null.
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 states a specific verb and resource: search nomado24's remote/hybrid job board. It clearly distinguishes this search tool from sibling tools get_job and get_job_statistics, which serve lookup and aggregate statistics rather than search.
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 explains which argument combinations select the legacy v1 versus v2 contract, that page plus structured filters is refused, and that cursor must be sent alone. It also directs agents to get_job for provenance, giving a clear alternative when per-field dataOrigin is needed.
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.
search_jobs1 field changedInput schema / properties / include_total / descriptionPrevious value: -"Whether to compute a total. 'exact' returns a number only when the database saw the whole predicate. With a post-projection filter active (skills, seniority, salary_min/salary_max, employment_type=unknown) the total is withheld: total is null and totalRelation is 'not_computed', because the base count would answer a different question than the one asked. Only at the count ceiling does 'exact' downgrade to 'estimated'."New value: +"Requested total mode (none, estimated, exact). Totals are currently withheld for every structured search: a current-text on-site-only serving gate is not visible to the database count, even without a named filter. The response reports total: null and totalRelation: 'not_computed' for all three modes."search_jobs4 fields changedInput schema / properties / include_total / descriptionPrevious value: -"Whether to compute a total. 'exact' downgrades itself to 'estimated' when a post-projection filter is active or the count ceiling is hit"New value: +"Whether to compute a total. 'exact' returns a number only when the database saw the whole predicate. With a post-projection filter active (skills, seniority, salary_min/salary_max, employment_type=unknown) the total is withheld: total is null and totalRelation is 'not_computed', because the base count would answer a different question than the one asked. Only at the count ceiling does 'exact' downgrade to 'estimated'."Input schema / properties / work_arrangement / descriptionPrevious value: -"Work arrangement. 'unknown' means the field is absent in the response, not a stored value"New value: +"Work arrangement. The board serves remote and hybrid postings only, so these are the only two values"Input schema / properties / work_arrangement / items / enumPrevious value: -[
- "remote",
- "hybrid",
- "remote_optional",
- "unknown"
-]New value: +[
+ "remote",
+ "hybrid"
+]Input schema / properties / work_arrangement / maxItemsPrevious value: -4New value: +2get_jobsearch_jobs17 fields changedInput schema / properties / applicant_regionAdded value: +{
+ "description": "Where a CANDIDATE may be: WORLDWIDE, EU or an ISO 3166-1 alpha-2 code. EEA/DACH/EMEA are refused, nomado24 does not classify them",
+ "items": {
+ "maxLength": 16,
+ "type": "string"
+ },
+ "maxItems": 20,
+ "type": "array"
+}Input schema / properties / companyAdded value: +{
+ "description": "Company name, exact match, case-insensitive. Aggregated postings carry no verified employer identity",
+ "items": {
+ "maxLength": 128,
+ "type": "string"
+ },
+ "maxItems": 10,
+ "type": "array"
+}Input schema / properties / countryAdded value: +{
+ "description": "Where the ROLE sits, ISO 3166-1 alpha-2 uppercase (DE, US)",
+ "items": {
+ "maxLength": 8,
+ "type": "string"
+ },
+ "maxItems": 20,
+ "type": "array"
+}Input schema / properties / cursorAdded value: +{
+ "description": "Continuation cursor from pagination.nextCursor. Send it ALONE (with per_page at most), never alongside filters: the cursor already carries the filter set it traverses",
+ "maxLength": 4096,
+ "type": "string"
+}Input schema / properties / employment_typeAdded value: +{
+ "description": "Employment type. 'unknown' means the field is absent in the response, not a stored value",
+ "items": {
+ "enum": [
+ "full_time",
+ "part_time",
+ "contract",
+ "freelance",
+ "temporary",
+ "internship",
+ "working_student",
+ "apprenticeship",
+ "minijob",
+ "volunteer",
+ "other",
+ "unknown"
+ ],
+ "type": "string"
+ },
+ "maxItems": 12,
+ "type": "array"
+}Input schema / properties / include_totalAdded value: +{
+ "description": "Whether to compute a total. 'exact' downgrades itself to 'estimated' when a post-projection filter is active or the count ceiling is hit",
+ "enum": [
+ "none",
+ "estimated",
+ "exact"
+ ],
+ "type": "string"
+}Input schema / properties / published_afterAdded value: +{
+ "description": "RFC 3339 date-time, e.g. 2026-08-01T00:00:00Z",
+ "maxLength": 64,
+ "type": "string"
+}Input schema / properties / salary_currencyAdded value: +{
+ "description": "ISO 4217. Only EUR: every salary is normalized to EUR at ingest and no conversion is performed",
+ "maxLength": 8,
+ "type": "string"
+}Input schema / properties / salary_maxAdded value: +{
+ "description": "Upper bound in normalized ANNUAL EUR (gross)",
+ "minimum": 0,
+ "type": "number"
+}Input schema / properties / salary_minAdded value: +{
+ "description": "Lower bound in normalized ANNUAL EUR (gross). Matches jobs whose salary interval overlaps it",
+ "minimum": 0,
+ "type": "number"
+}Input schema / properties / seniorityAdded value: +{
+ "description": "Seniority. 'unknown' means the field is absent in the response, not a stored value",
+ "items": {
+ "enum": [
+ "student",
+ "entry",
+ "junior",
+ "mid",
+ "senior",
+ "lead",
+ "manager",
+ "director",
+ "executive",
+ "unknown"
+ ],
+ "type": "string"
+ },
+ "maxItems": 10,
+ "type": "array"
+}Input schema / properties / skillsAdded value: +{
+ "description": "Skill slugs, lowercase kebab-case (react, node-js)",
+ "items": {
+ "maxLength": 64,
+ "type": "string"
+ },
+ "maxItems": 20,
+ "type": "array"
+}Input schema / properties / skills_matchAdded value: +{
+ "description": "How to combine skills. Requires skills",
+ "enum": [
+ "any",
+ "all"
+ ],
+ "type": "string"
+}Input schema / properties / sortAdded value: +{
+ "description": "One of published_desc, verified_desc, salary_asc, salary_desc. Default published_desc",
+ "maxLength": 32,
+ "type": "string"
+}Input schema / properties / sourceAdded value: +{
+ "description": "Feed/source ids, lowercase kebab-case",
+ "items": {
+ "maxLength": 64,
+ "type": "string"
+ },
+ "maxItems": 20,
+ "type": "array"
+}Input schema / properties / verified_afterAdded value: +{
+ "description": "RFC 3339 date-time. Compares against the SAME value the response reports as lastVerifiedAt",
+ "maxLength": 64,
+ "type": "string"
+}Input schema / properties / work_arrangementAdded value: +{
+ "description": "Work arrangement. 'unknown' means the field is absent in the response, not a stored value",
+ "items": {
+ "enum": [
+ "remote",
+ "hybrid",
+ "remote_optional",
+ "unknown"
+ ],
+ "type": "string"
+ },
+ "maxItems": 4,
+ "type": "array"
+}search_jobs1 field changedInput schema / properties / per_page / maximumPrevious value: -9007199254740991New value: +25get_job_statisticssearch_jobsJobs from 10,000+ company career sites, Germany's Arbeitsagentur, Wellfound startups; EU suppliers.
Live job postings worldwide: search by role, country, remote, salary, visa; pay and demand stats.
Remote jobs: job search, salary data and application tracking, or post a job and hire candidates
Search remote jobs from company career pages by title, location, seniority, salary, and tech stack.
Add one secure layer between your agents and this server.