skills.rest
davila7
davila7COMMUNITY
Explore all skills by this author →
Website𝕏@dani_avila7GitHub

workorai

Search jobs and hire candidates through the WorkorAI MCP talent marketplace.

30.5k|3.5k|Updated Jul 4, 2025
Search Tags:#mcp#job search#hiring#recruiting#candidate matching#workorai#talent marketplace
One-click install
npx skills add https://github.com/davila7/claude-code-templates --skill workorai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: workorai
Source: https://github.com/davila7/claude-code-templates/tree/main/cli-tool/components/skills/career/workorai
Command: npx skills add https://github.com/davila7/claude-code-templates --skill workorai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Job seekers and recruiters waste hours keyword-scraping job boards with no insight into why a role or candidate actually fits. This Skill connects an AI agent to the WorkorAI talent marketplace via its MCP server, enabling semantic job matching for candidates and white-box, evidence-backed candidate evaluation for employers.

Core Features & Use Cases

  • Candidate job search: Semantically ranked vacancy search with per-job match scores, matched/missing skill breakdowns, and one-click apply, invitation accept/decline, withdraw, and saved-job management across 9 candidate tools.
  • Employer hiring pipeline: Post jobs from free-text descriptions, search candidates by vacancy or free-form query with tiered match bands (best/good/weak), inspect interview evidence and Q&A, invite candidates, and manage the applicant review funnel across 19 employer tools.
  • Secure credential handling: A cross-platform credential-store script saves WorkorAI MCP keys to the OS keychain (macOS Keychain, Linux Secret Service, Windows SecretManagement) with role-scoped slots and redacted output.
  • Use Case: A candidate says "find me a remote senior backend job" — the agent retrieves their saved key, runs a ranked search, presents an Agent Pick with a fit breakdown from the match explanation, and applies on confirmation.

Quick Start

Ask the agent to find you a job on WorkorAI or to find candidates for your vacancy, and provide your WorkorAI MCP key when prompted.

Frequently Asked Questions about workorai

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search for jobs with the WorkorAI MCP server?▼

Provide your WorkorAI MCP key (wai_...) from the candidate MCP page, then the agent calls candidate.search_jobs with optional filters like workModel, seniority, and jobType. Results are semantically ranked against your evaluated profile interview with per-job match explanations.

How do employers find and evaluate candidates on WorkorAI?▼

Employers call employer.search_candidates_for_job with a jobId and tier 'best' to get ranked candidates with match explanations, then use employer.get_candidate_evidence for interview facts and Q&A before inviting via employer.invite_candidate. An employer-scoped MCP key from the employer dashboard is required.

Where do I get a WorkorAI MCP API key?▼

Candidates generate a key at workorai.com/candidate/home?tab=mcp after completing a profile and an evaluated interview. Employers generate one from the Employer MCP card on the employer dashboard. Keys use the wai_ prefix and the role is determined server-side.

Why does candidate.apply_to_job return GATE_LOCKED or GATE_EVALUATING?▼

Applying requires a completed and evaluated profile interview. GATE_LOCKED means no interview is complete, GATE_EVALUATING means evaluation is still running, and GATE_FAILED means you must retake the interview. Finish the onboarding steps instead of retrying the call.

How are WorkorAI MCP keys stored securely?▼

The credential-store.mjs script saves keys to the OS secret store (macOS Keychain, Linux Secret Service via secret-tool, or Windows PowerShell SecretManagement) with separate candidate and employer slots. A 0600-permission shared file fallback is available only with explicit consent, and keys are redacted in all output.

Why does search_jobs return results with no match score?▼

A free-text query or a candidate without a completed interview triggers a recency-based browse instead of semantic ranking, so matchScore is null. Complete the profile interview and use structured filters instead of free text to get ranked results with fit explanations.

Related Skills

Explore More Product & Management Skills→
2nth-ai2nth-ai
1

biz/hr/recruitment

Automate recruitment from job descriptions to offer letters with ATS integration.

Official
Advanced
adrianliechtiadrianliechti
4

recruiting-pipeline

Track candidate statuses across recruiting stages from sourcing to offer acceptance.

Community
Intermediate
lzlatkinlzlatkin

EM Interview Prep

Generate tailored Engineering Manager interview preparation documents from resumes and historical data.

Community
Advanced
RoxabiRoxabi
9

linkedin-apply

Scrape LinkedIn job postings and score them against a candidate profile.

Official
Advanced