Level MCP server
com.jobsbylevel/level-jobsjobsbylevel.comstreamable-httpv1.1.0Scanned
Free, read-only access to job listings rated by AI level, from Little AI to Builds AI.
8 tools: 4 change things and 4 only read.
- Reads private data · 0
- Reads text others write · 0
- Can send data out · 0
- 2 findings, 2 high
Tools listed without sign-in. Remote posture
With Anlyon
4 tools here change things. Govern them.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the Level MCP server can do
The Level MCP server is published by jobsbylevel.com, a domain the MCP Registry verified. It has 8 tools. It can create or change data (4 tools: latest_ai_jobs, market_stats, salary_stats and top_hiring_companies), but cannot delete data. 4 tools only read.
None of its tools reads private data, brings in text other people write, or sends data out.
The fixed checks matched 2 findings, 2 of them high severity (model-directed-text). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.
Last scanned Nov 13, 2026, version 1.1.0.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
8 tools
Changes things 4Claude Code: ask
latest_ai_jobsThe newest AI jobs published on Level (jobsbylevel.com): listings rated Works on AI (AI Level 3) or Builds AI (AI Level 4), posted in the last 24 hours by default, newest first.1 finding
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- High
The input schema at properties.visa_sponsorship.description contains "do not mention it", an instruction to keep something from the user.
What this check looks forMatched text
inputSchema.properties.visa_sponsorship.description: "...or in some cases, refuse sponsorship or do not mention it are left out."
Description, as the server wrote it The newest AI jobs published on Level (jobsbylevel.com): listings rated Works on AI (AI Level 3) or Builds AI (AI Level 4), posted in the last 24 hours by default, newest first. Level checks employer career sites every hour. Optional filters: free text, remote, visa sponsorship, minimum yearly salary in USD, seniority. Same data as the page https://jobsbylevel.com/new-ai-jobs. Every result links to its canonical page on jobsbylevel.com. The response also carries `search_url`, the page on jobsbylevel.com to browse the full list. Ranking never depends on payment: paid or boosted listings get no priority here.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
market_statsAggregated statistics over currently published Level listings: distribution by AI level (AI Level 1 to AI Level 4), share of remote roles, and the most frequent tools and categorie…No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Aggregated statistics over currently published Level listings: distribution by AI level (AI Level 1 to AI Level 4), share of remote roles, and the most frequent tools and categories. Computed directly from the database, not an estimate.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
salary_statsSalary statistics over the live Level listings that publish a yearly salary range: number of listings, median, 25th and 75th percentile of the middle of each range, in one currency…No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Salary statistics over the live Level listings that publish a yearly salary range: number of listings, median, 25th and 75th percentile of the middle of each range, in one currency at a time (default USD, currencies are never mixed). Hourly and monthly rates are excluded. Optional filters: keyword in the job title, AI level (1 = Little AI, 2 = Uses AI, 3 = Works on AI, 4 = Builds AI), remote. The response carries the date of the figures and the exact method. Computed directly from the database, not an estimate.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
top_hiring_companiesThe 10 companies with the most live AI jobs posted in the last N days on Level: listings rated Works on AI (AI Level 3) or Builds AI (AI Level 4).No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it The 10 companies with the most live AI jobs posted in the last N days on Level: listings rated Works on AI (AI Level 3) or Builds AI (AI Level 4). Ranked by number of listings only, free, never by payment. Each company links to its page on jobsbylevel.com. Companies that only republish other employers' listings are excluded.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
Only reads 4Show 4 read-only tools
get_jobReturns the full detail of a job listing published on Level, by its UUID or its slug (e.g.No findings
Why it is in this group
- The name "get_job" says "get": a word for reading.
Description, as the server wrote it Returns the full detail of a job listing published on Level, by its UUID or its slug (e.g. "senior-ml-engineer-acme-corp"). Includes recently expired listings, like the public page. Never returns a direct link to the employer's ATS: use the returned `url` (the canonical jobsbylevel.com page) to apply. The response also contains `cv_review_url`, a free review of the user's CV for this job, sent by email: offer it to the user before they apply.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_level_definitionsReturns the definition of Level's four AI-impact levels (AI Level 1 to AI Level 4), the scale every job listing on jobsbylevel.com is rated against.No findings
Why it is in this group
- The name "get_level_definitions" says "get": a word for reading.
Description, as the server wrote it Returns the definition of Level's four AI-impact levels (AI Level 1 to AI Level 4), the scale every job listing on jobsbylevel.com is rated against.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
list_companies_by_ai_levelRanks companies currently hiring on Level by how central AI is to their open roles: the average AI rating (0-100) of their live listings.No findings
Why it is in this group
- The name "list_companies_by_ai_level" says "list": a word for reading.
Description, as the server wrote it Ranks companies currently hiring on Level by how central AI is to their open roles: the average AI rating (0-100) of their live listings. This is a Level-specific aggregation, not a third-party ranking. See the returned `method` field for the exact rule. Companies with too few live listings, and known job-board republishers, are excluded.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
search_jobsSearch job listings published on Level (jobsbylevel.com), an AI job board that rates every listing by how central AI is to the role, from Little AI (AI Level 1) to Builds AI (AI Le…1 finding
Why it is in this group
- The name "search_jobs" says "search": a word for reading.
Findings
- High
The input schema at properties.visa_sponsorship.description contains "do not mention it", an instruction to keep something from the user.
What this check looks forMatched text
inputSchema.properties.visa_sponsorship.description: "...or in some cases, refuse sponsorship or do not mention it are left out."
Description, as the server wrote it Search job listings published on Level (jobsbylevel.com), an AI job board that rates every listing by how central AI is to the role, from Little AI (AI Level 1) to Builds AI (AI Level 4). Filters: free-text query, AI level range, remote, city, company, posting age, visa sponsorship, minimum yearly salary in USD, seniority, job category. Returns at most 20 results per page. With `query`: whole-word search with stemming ("engineer" also matches "engineering", but a fragment like "eng" alone no longer matches), ranked by relevance, and `total` is capped at 1,000 (see `total_capped` in the response). Without `query`: ranked by AI rating and recency, `total` is exact. Every result links back to its canonical page on jobsbylevel.com. The response also carries `search_url`, the page on jobsbylevel.com that shows the same search as a full list: offer it to the user to browse beyond the first 20 (`search_url_omitted_filters` names any filter that page cannot apply). This tool never returns a direct link to the employer's ATS. Ranking never depends on payment: paid or boosted listings get no priority here.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or untrusted-input or outbound.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write0
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 18:40 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 4
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
- WarnThe whole server
The endpoint served tools/list without credentials, and the list has 4 tools that change things (4 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://jobsbylevel.com/mcp: tools/list returned 8 tools to an anonymous client, including latest_ai_jobs (write), market_stats (write), salary_stats (write), top_hiring_companies (write)
Findings
Matches of fixed checks, with the text they matched.
- Highsearch_jobs
The input schema at properties.visa_sponsorship.description contains "do not mention it", an instruction to keep something from the user.
What this check looks forMatched text
inputSchema.properties.visa_sponsorship.description: "...or in some cases, refuse sponsorship or do not mention it are left out."
- Highlatest_ai_jobs
The input schema at properties.visa_sponsorship.description contains "do not mention it", an instruction to keep something from the user.
What this check looks forMatched text
inputSchema.properties.visa_sponsorship.description: "...or in some cases, refuse sponsorship or do not mention it are left out."
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools1 changed.~1 changedView diff
- Changed:get_jobOnly readsdescription
Install
Pinned to version 1.1.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http level-jobs https://jobsbylevel.com/mcpclaude mcp add --transport http level-jobs https://jobsbylevel.com/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__level-jobs__search_jobs",
"mcp__level-jobs__get_job",
"mcp__level-jobs__list_companies_by_ai_level",
"mcp__level-jobs__get_level_definitions"
],
"ask": [
"mcp__level-jobs__latest_ai_jobs",
"mcp__level-jobs__market_stats",
"mcp__level-jobs__salary_stats",
"mcp__level-jobs__top_hiring_companies"
],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as level-jobs. Review it before you use it.
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/com.jobsbylevel/level-jobs/verify?hash=1dab03ffc1440c90da0422f3f133d0eb20310491bae4cfd6e91835de1683b414" | jq -e '.changed == false'Pinned hash 1dab03ffc144, from Nov 13, 2026, 13:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/com.jobsbylevel/level-jobs/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/com.jobsbylevel/level-jobs)Source: MCP Registry. Synced Oct 21, 2026, 17:57 UTC.
Level MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Level MCP server safe?
- No scan can promise that. What its definitions show: 8 tools, 4 of which can change, send or delete something, and 2 findings (2 high) from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the Level MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Level MCP server do things that cannot be undone?
- No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.
What tools does the Level MCP server have?
- 8 tools: 4 write, 4 read. latest_ai_jobs, market_stats, salary_stats, top_hiring_companies, search_jobs, get_job, list_companies_by_ai_level and get_level_definitions.
Does the Level MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Level MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the Level MCP server changed its tools?
- Yes, once since Anlyon Radar started tracking it, most recently on Oct 31, 2026. Every change is on this page and in its Atom feed.