job-ad-intelligence-mcp-production.up.railway.app

Classify the seniority level of a job from its title and optional description. Returns one of: entry, junior, mid, senior, lead, head_of, executive, unknown. Also returns the signals that triggered the classification and a confidence score 0–1. Uses title signals first (highest weight) then description signals.

Parse a freeform salary string into a structured min/max/currency/period object. Handles formats including £45k–£60k, up to $120,000, €80,000 pa, from £500 per day, $120k, competitive, DOE, negotiable. Returns nulls and explanatory notes for ambiguous or non-numeric inputs. Supports GBP, USD, and EUR.

Extract structured data from a job advert. Pass raw text or a public URL. Returns title, company, location, remote policy, employment type, salary range, seniority level, required and preferred skills, responsibilities, requirements, benefits, and application instructions as JSON.

Generate 5–10 targeted questions a candidate should ask before applying or during a screening call. Questions are tailored to gaps, ambiguities, and notable claims in the job advert — not generic filler. Each question includes a "why it matters" explanation and is categorised by topic (role, company, compensation, flexibility, process, expectations).

Score a candidate CV against a job advert using skills, experience, tools, and seniority signals. Returns an overall score 0–100, matched and missing skills, experience alignment, red flags, and interview talking points. Skills-based comparison only — no protected-characteristic inference. Includes a mandatory disclaimer for responsible use in hiring contexts.