hunaras-recruiting

Hunaras Recruiting

AI-native recruiting platform connecting employers with candidates through AI-powered interviews.

Overview

Hunaras is a dual-role recruiting platform where:

• Employers post jobs, search a pre-assessed talent pool, review AI interview results, and schedule meetings
• Candidates find jobs, take AI video interviews, receive scored assessments, and manage their career

All interactions happen through natural language via MCP tools. The server authenticates with an API key (hnar_ prefix) set in HUNARAS_API_KEY, and tools are scoped by role -- employer users see employer tools, candidate users see candidate tools. The shared get_market_insights tool is available to both roles.

Configuration

{
  "mcpServers": {
    "hunaras": {
      "command": "node",
      "args": ["path/to/mcp-server/dist/index.js"],
      "env": {
        "HUNARAS_API_KEY": "hnar_your_api_key_here",
        "HUNARAS_API_URL": "https://hunaras.com"
      }
    }
  }
}

Quick Start

For Candidates

I want to...	Tool(s)
Find a job	search_jobs then get_job_details
Apply to a job	apply_to_job
See my application status	get_my_applications
Check interview scores	get_my_assessments then get_assessment_details
Update my profile	update_profile
Import from LinkedIn	sync_linkedin
Show/hide from employers	update_visibility
Respond to a meeting invite	get_my_appointments then respond_to_appointment
Rate an employer	get_pending_ratings then rate_employer
Check notifications	get_my_notifications

For Employers

I want to...	Tool(s)
Post a job	create_job
See my listings	list_my_jobs
Find candidates on platform	search_candidates then get_candidate_profile
Source from LinkedIn	hunt_candidates then invite_candidate
Check market data	get_market_insights

Candidate Tools

1. Find and Apply for a Job

Flow: search_jobs -> get_job_details -> apply_to_job

Use search_jobs with a natural language query. Results include a semantic match percentage. Drill into a job with get_job_details using the job ID. Apply with apply_to_job, optionally including a cover letter.

Tool reference:

• search_jobs -- Search for jobs using AI semantic search

  • query (string, required): Natural language query, e.g. "frontend developer with React experience"
  • location (string, optional): Filter by location, e.g. "Remote", "New York"
  • type (enum, optional): full-time | part-time | contract | internship
  • limit (number, optional, default 10): Max results (1-50)

• get_job_details -- Get full details of a job posting

  • job_id (string, required): The job ID from search results

• apply_to_job -- Submit an application

  • job_id (string, required): The job ID to apply to
  • cover_letter (string, optional): Cover letter text

2. Check Interview Results

Flow: get_my_assessments -> get_assessment_details

After completing an AI interview, view your scored assessment. Assessments rate you across 5 dimensions on a 0-100 scale:

• Technical Skills -- domain knowledge and technical depth
• Communication -- clarity, articulation, and professionalism
• Problem Solving -- analytical thinking and structured approach
• Experience -- relevant background and accomplishments
• Culture Fit -- alignment with team values and work style

Score interpretation:

• 80-100: Exceptional
• 60-79: Strong
• 40-59: Average
• Below 40: Below expectations

Each assessment also includes strengths, areas for improvement, and an overall recommendation.

Tool reference:

• get_my_assessments -- List all your AI interview assessments with scores

  • No parameters

• get_assessment_details -- Get a full detailed assessment report

  • assessment_id (string, required): The assessment ID

3. Set Up Your Profile

Flow: get_my_profile -> update_profile -> sync_linkedin -> update_visibility

A complete profile increases your visibility to employers. Start by reviewing your current profile, then update fields. Optionally sync from LinkedIn to import experience and skills automatically.

Tool reference:

• get_my_profile -- View your current profile, scores, and visibility status

  • No parameters

• update_profile -- Update profile fields (only provided fields change)

  • headline (string, optional): Professional title, e.g. "Senior Frontend Developer"
  • bio (string, optional): Brief bio about yourself
  • location (string, optional): Your location, e.g. "San Francisco, CA"
  • skills (array of strings, optional): Your skills, e.g. ["React", "TypeScript", "Node.js"]

• sync_linkedin -- Sync your profile with LinkedIn data

  • linkedin_url (string, optional): LinkedIn profile URL, e.g. "https://linkedin.com/in/username". Omit to use your linked account.

• update_visibility -- Toggle talent pool visibility

  • visible (boolean, required): true to appear in employer searches, false to hide

4. Manage Meetings

Flow: get_my_appointments -> respond_to_appointment

When an employer wants to schedule a meeting after reviewing your assessment, you receive an appointment proposal. You can accept, reschedule with a new time, or decline with a reason.

Tool reference:

• get_my_appointments -- List meeting proposals from employers

  • No parameters

• respond_to_appointment -- Accept, reschedule, or reject a meeting

  • appointment_id (string, required): The appointment ID
  • action (enum, required): accept | reschedule | reject
  • proposed_at (string, optional): New proposed time (ISO 8601) -- required for reschedule
  • message (string, optional): Message to employer
  • rejection_reason (string, optional): Reason for rejection

5. Rate Employers

Flow: get_pending_ratings -> rate_employer -> get_my_ratings

After completing an interview process, you can rate the employer on their hiring experience. This helps other candidates.

Tool reference:

• get_pending_ratings -- See which employers you can rate

  • No parameters

• rate_employer -- Submit a rating for an employer

  • job_id (string, required): Job ID to rate the employer for
  • overall_score (number, required): Rating 1-5 stars
  • trigger (enum, required): AFTER_MEETING | AFTER_JOB_CLOSED | AFTER_AI_INTERVIEW
  • comment (string, optional): Written feedback (min 20 chars if provided)
  • dimensions (object, optional): Dimension scores, e.g. {"communication": 4, "professionalism": 5}

• get_my_ratings -- View all ratings you have received and given

  • No parameters

6. Check Inbox

Flow: get_my_notifications -> mark_notifications_read

Stay on top of application updates, meeting invitations, and other activity.

Tool reference:

• get_my_notifications -- List your notifications

  • unread_only (boolean, optional): Filter to unread notifications only

• mark_notifications_read -- Mark notifications as read

  • notification_id (string, optional): Specific notification ID, or omit to mark all as read

7. Track Applications

Flow: get_my_applications

View all your submitted applications with their current status.

Application statuses:

• pending -- submitted, awaiting review
• screening -- employer is reviewing
• interviewed -- AI interview completed
• shortlisted -- you made the short list
• appointment -- meeting scheduled
• hired -- congratulations!
• rejected -- not selected for this role

Tool reference:

• get_my_applications -- List all applications with statuses
  • No parameters

Employer Tools

1. Post a Job and Hire

Flow: create_job -> wait for applications -> get_candidate_profile

Post a job with a title, description, and requirements. Candidates apply and take an AI interview. Review their profiles and assessment scores. Free tier allows 2 active jobs and 1 candidate reveal; Pro unlocks unlimited.

Tool reference:

• create_job -- Post a new job listing

  • title (string, required): Job title, e.g. "Senior Frontend Engineer"
  • description (string, required): Full job description (markdown supported)
  • requirements (array of strings, optional): List of requirements
  • location (string, optional, default "Remote"): Job location
  • type (enum, optional, default "full-time"): full-time | part-time | contract | remote
  • salary (string, optional): Salary range, e.g. "$120k-$160k"
  • question_count (number, optional, default 8): Interview questions (3-20)

• list_my_jobs -- List your job postings with application counts

  • include_inactive (boolean, optional, default false): Include closed/draft jobs

• get_candidate_profile -- View a candidate's full profile and assessment (counts as a reveal)

  • candidate_id (string, required): Candidate user ID from search results

2. Search the Talent Pool

Flow: search_candidates -> get_candidate_profile -> invite_candidate

Search the pool of pre-assessed candidates using natural language. Filter by location and minimum assessment score. This is the fastest way to find qualified candidates since they already have AI interview scores.

Requires Pro subscription.

Tool reference:

• search_candidates -- Semantic search across the talent pool
  • query (string, required): Natural language query, e.g. "senior React developer with AWS experience"
  • location (string, optional): Filter by location
  • min_score (number, optional): Minimum assessment score (1-10 scale)
  • limit (number, optional, default 10): Max results (1-50)

3. Source from LinkedIn

Flow: hunt_candidates -> review results -> invite_candidate

Search LinkedIn for external candidates who are not yet on Hunaras. Found candidates can be invited to join the platform. Hunaras handles the outreach on your behalf.

Requires Pro subscription. Maximum 20 invites per day.

Tool reference:

• hunt_candidates -- Search LinkedIn for external candidates

  • query (string, required): Search query, e.g. "React developer San Francisco"
  • keywords (array of strings, optional): Additional keywords to refine results

• invite_candidate -- Invite an external candidate to Hunaras

  • linkedin_url (string, required): Full LinkedIn profile URL
  • name (string, required): Candidate name
  • headline (string, optional): Their LinkedIn headline or current role
  • company (string, optional): Their current company

4. Market Research

Flow: get_market_insights

Get market data to inform your hiring strategy -- active job counts, candidate pool size, job type distribution, top locations, and salary ranges.

Tool reference:

• get_market_insights -- Get job market data (available to both roles)
  • category (string, optional): Filter by job category, e.g. "engineering", "marketing"
  • location (string, optional): Filter by location, e.g. "Remote", "New York"

5. Manage Your Pipeline

Flow: list_my_jobs -> review statuses and applicant counts

Monitor your job postings, see how many applications each has received, and track which positions need attention.

Decision Trees

For Employers: Talent Pool vs LinkedIn Hunt

Criteria	Talent Pool (search_candidates)	LinkedIn (hunt_candidates)
Candidates pre-assessed?	Yes -- AI interview scores ready	No -- must complete interview
Speed to hire	Fast -- profiles ready to review	Slower -- outreach + onboarding
Reach	Platform users only	All of LinkedIn
Cost	Pro subscription	Pro subscription + daily limits
Best for	Active hiring with known criteria	Hard-to-fill or niche roles

Recommendation: Start with search_candidates. Fall back to hunt_candidates when the talent pool does not have enough matches.

For Candidates: Visibility Strategy

• Set visible (update_visibility with visible: true) if you are actively looking or open to opportunities. Employers with Pro can discover you.
• Set hidden (update_visibility with visible: false) if you are not looking. Your applications and assessments are unaffected.
• Complete your profile before going visible -- employers see your headline, bio, skills, and assessment scores.

Free vs Pro (Employers)

Feature	Free	Pro ($200/month)
Active job postings	2	Unlimited
Candidate reveals	1	Unlimited
Talent pool search	Not available	Full semantic search
LinkedIn sourcing	Not available	20 invites/day
Meeting scheduling	Not available	Full access
Team seats	1	Multiple

Candidates always use Hunaras for free.

Handling Upgrade Prompts

When a tool returns an upgrade-required message, inform the user:

• What feature they tried to use
• That it requires a Pro subscription
• Direct them to https://hunaras.com/pricing

Error Handling

Status	Meaning	Suggested Response
401	Not authenticated	"Please check your API key configuration"
403	Forbidden / upgrade needed	Check if upgradeRequired -- suggest Pro upgrade at hunaras.com/pricing
404	Resource not found	Verify the ID/slug is correct and try again
409	Duplicate action	User already performed this action (e.g., already applied, already rated)
429	Rate limited	Daily limit reached -- wait or try again tomorrow
502	Upstream service down	"This service is temporarily unavailable, please try again later"

Domain Tips

Writing Job Descriptions (create_job)

• Be specific about requirements vs nice-to-haves
• Include a salary range -- jobs with listed salary get more qualified applicants
• Specify remote/hybrid/onsite clearly in the location field
• Use the requirements array for structured, scannable requirements
• Set question_count between 5-12 for most roles; use higher counts (15-20) for senior/complex positions

Interpreting Assessments

• Dimension breakdown matters more than the overall score
• Check strengths and weaknesses for a nuanced view beyond the numbers
• The recommendation field gives an AI-generated hiring recommendation
• Scores are calibrated relative to the job requirements, not absolute

Candidate Profile Optimization

• A complete profile (headline, bio, skills, location) ranks higher in talent pool search
• Syncing LinkedIn imports verified experience data
• Multiple assessments from different job applications show range and consistency

Reference Files

For detailed multi-step orchestration patterns, see:

• Employer Workflows -- complete employer journey guides
• Candidate Workflows -- complete candidate journey guides