Dice MCP server
io.github.DiceHoldingsInc/dice-mcpDiceHoldingsIncstreamable-httpv1.0.0Scanned
Dice MCP: Search for tech jobs on Dice.com
3 tools: 3 only read.
- Reads private data · 0
- Reads text others write · 0
- Can send data out · 0
- 1 finding
Tools listed without sign-in. Remote posture
With Anlyon
Decide which of this server's tools your agent can use.
- 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 Dice MCP server can do
The Dice MCP server is published by the GitHub account DiceHoldingsInc. It has 3 tools. Every one of them only reads: going by their definitions, none of them creates, changes, sends or deletes anything.
None of its tools reads private data, brings in text other people write, or sends data out.
The fixed checks matched 1 finding (long-description). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 16, 2026, version 1.0.0.
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.
3 tools
Only reads 3Show 3 read-only tools
get_companyRetrieve the name and description of the company behind a job, by job ID.No findings
Why it is in this group
- The name "get_company" says "get": a word for reading.
Description, as the server wrote it Retrieve the name and description of the company behind a job, by job ID. Use this tool when users want more detail about a company than search_jobs/get_job_details return (e.g. after finding a job and wanting to know more about the company that posted it). DO NOT use for: searching/discovering companies or jobs by keyword (use search_jobs instead). LLM USAGE INSTRUCTIONS: - The typical flow is: search_jobs to find jobs -> user picks one -> offer to look up the company -> if the user agrees, call this tool with that same job_id (the same guid used for get_job_details; no separate lookup step is needed first). - Do not call this automatically right after get_job_details; ask the user first unless they already explicitly asked for company details. - Not every company has published an employer branding profile, and some profiles are marked not visible by the company. When that happens this tool does NOT raise — it returns a Company with `message` set and empty name/desc. Relay that message to the user (e.g. "Looks like that company hasn't published a public profile.") rather than treating it as an error. Args: job_id: The unique identifier of the job whose company to look up (required). This must be the job's `guid` field from a search_jobs result, NOT its `id` field — `id` is a different, internal identifier that this tool does not accept. Returns: Company: Contains: - name: The company name (empty if message is set) - desc: The company description (empty if message is set) - message: Set instead of name/desc when there is no visible company profile to show; None when name/desc are populated Raises: Exception: If the job is not found, an API call itself fails (network error, non-200 response, GraphQL errors), or input validation errors occur
Annotations the server declares
- title: Get Company
- readOnlyHint: true
- destructiveHint: false
get_job_detailsRetrieve a job's full description and associated skills by job ID.No findings
Why it is in this group
- The name "get_job_details" says "get": a word for reading.
Description, as the server wrote it Retrieve a job's full description and associated skills by job ID. Use this tool when users want more detail about a specific job than search_jobs returns (e.g. after finding a job via search_jobs and wanting its full description and skills list). DO NOT use for: searching/discovering jobs by keyword (use search_jobs instead). LLM USAGE INSTRUCTIONS: - Only call this after the user has picked out a specific job from search_jobs results (or otherwise supplied a job's guid directly) — do not call it speculatively for every search result. - After returning job details, do NOT automatically call get_company. Instead, offer to look up the company (e.g. "Want me to pull up more info on the company?") and only call get_company, passing this same job_id, if the user says yes. Args: job_id: The unique identifier of the job to retrieve (required). This must be the job's `guid` field from a search_jobs result, NOT its `id` field — `id` is a different, internal identifier that this tool does not accept. Returns: JobDetails: Contains: - description: The full job description - skills: List of Skill objects associated with the job Raises: Exception: If the API call fails, the job is not found, or input validation errors occur
Annotations the server declares
- title: Get Job Details
- readOnlyHint: true
- destructiveHint: false
search_jobsSearch for job listings by keyword, location, and filters.1 finding
Why it is in this group
- The name "search_jobs" says "search": a word for reading.
Findings
- Info
The description is 3,985 characters long, over the 2,000 this check looks for.
What this check looks forMatched text
description ends: "...sults Raises: Exception: If API call fails or input validation errors occur"
Description, as the server wrote it Search for job listings by keyword, location, and filters. Returns job details, company info, and application links. Use this tool when users want to find jobs, search employment opportunities, or explore job openings. DO NOT use for: applying to jobs, submitting applications, or making employment decisions. LLM USAGE INSTRUCTIONS: - ALWAYS provide the keyword parameter (required) - When presenting results to users, include BOTH the job details URL (detailsPageUrl) AND the company page URL (companyPageUrl) for each job - Use location to find geographically relevant positions - Combine filters to refine searches (e.g., workplace_types=['Remote'] for remote work) - Use posted_date to find recent openings ('ONE'=1 day, 'THREE'=3 days, 'SEVEN'=7 days) - Default jobs_per_page is reasonable, increase for comprehensive searches - Use company_name to scope results to a specific employer - Use sort='datePosted' when the user wants the newest listings instead of the most relevant - Use facets to get aggregate counts (e.g., how many results are remote vs on-site) without paging through all results - After presenting results, do NOT automatically call get_job_details for every job returned. Wait for the user to indicate which specific job(s) they're interested in, then pass that job's guid to get_job_details. IMPORTANT - AI DISCLOSURE REQUIREMENT: When presenting job search results to users, you MUST include an appropriate disclosure that these results were retrieved using AI assistance. Example disclosure language: "These job listings were found using AI-powered search. Please review all job details carefully and verify information directly with employers before applying." This tool provides job listing data only. Final employment decisions should always involve human judgment and direct review of complete job postings. Args: keyword: The job keyword or title to search for (required) location: Geographic location for the job search (city, state, country) radius: Search radius from the specified location (minimum 1, requires radius_unit) radius_unit: Unit for search radius. Options: 'mi', 'km' (requires radius) jobs_per_page: Number of jobs to return per page (1-100, defaults to 5) page_number: Page number for pagination (1-based, default is 1) sort: Sort order for results. Options: 'relevance', 'datePosted' (defaults to relevance) posted_date: Filter by posting date. Options: 'ONE' (1 day), 'THREE' (3 days), 'SEVEN' (7 days) workplace_types: Workplace arrangements. Options: 'Remote', 'On-Site', 'Hybrid' employment_types: Employment types. Options: 'FULLTIME', 'CONTRACTS', 'PARTTIME', 'THIRD_PARTY', 'INTERNSHIP' employer_types: Employer types. Options: 'Direct Hire', 'Recruiter', 'Other' willing_to_sponsor: Filter for employers willing to sponsor work authorization (boolean) easy_apply: Filter for jobs with easy application process (boolean) company_name: Filter by company name fields: Specific fields to include in response (optional, returns all exposed fields by default) facets: Facet dimensions to aggregate (optional). Options: 'employmentType', 'postedDate', 'workFromHomeAvailability', 'workplaceTypes', 'employerType', 'easyApply', 'isRemote', 'willingToSponsor' Returns: JobSearchResult: Contains: - data: List of JobDisplayFields with job details including: * guid: The job's identifier to pass as job_id to get_job_details for the full description and skills list (NOT the `id` field, which is a different, internal identifier not accepted by get_job_details) * detailsPageUrl: Direct link to full job posting * companyPageUrl: Link to company profile page * title, summary, salary, location, employmentType, etc. - metadata: Search metadata with pagination info and facet results Raises: Exception: If API call fails or input validation errors occur
Annotations the server declares
- title: Job Search
- readOnlyHint: true
- destructiveHint: 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, 18:57 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
- 0
- 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
Findings
Matches of fixed checks, with the text they matched.
- Infosearch_jobs
The description is 3,985 characters long, over the 2,000 this check looks for.
What this check looks forMatched text
description ends: "...sults Raises: Exception: If API call fails or input validation errors occur"
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 1.0.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http dice https://mcp.dice.com/mcpclaude mcp add --transport http dice https://mcp.dice.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__dice__search_jobs",
"mcp__dice__get_job_details",
"mcp__dice__get_company"
],
"ask": [],
"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 dice. 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/io.github.DiceHoldingsInc/dice-mcp/verify?hash=cb57e9e5ddb751d01b16d87cf6ebaeb14ba648f53691aba627fd1f306cc571e0" | jq -e '.changed == false'Pinned hash cb57e9e5ddb7, from Nov 16, 2026, 17:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.DiceHoldingsInc/dice-mcp/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.DiceHoldingsInc/dice-mcp)Source: MCP Registry. Synced Oct 21, 2026, 18:15 UTC.
Dice MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Dice MCP server safe?
- No scan can promise that. What its definitions show: 3 tools, all of which only read, and 1 finding from fixed checks. Read-only tools can still pass private data to a model.
Can the Dice MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Dice 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 Dice MCP server have?
- 3 tools: 3 read. search_jobs, get_job_details and get_company.
Does the Dice MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Dice MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the Dice MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 16, 2026.