candidate-dossierlisted

# Candidate Dossier (Claim Verification) ## Overview Build a complete, **evidence-cross-referenced** profile of one candidate by aggregating their full public footprint, then verifying each claim against real data. Output: verified strengths, red flags, key insights, and tailored interview questions — every line backed by a clickable source. Core principle: **a claim is only as good as the artifact that proves it.** Years, skills, and impact get checked against GitHub history, papers, launches, and answers — not taken at face value. ## When to Use - Vetting a candidate before reaching out or advancing them - A resume makes strong claims and you want them substantiated - Interview prep — you want sharp, evidence-driven questions - A candidate from `evidence-sourcing` looks promising and warrants a deep dive **Not for:** broad discovery (`evidence-sourcing`) or numeric scoring against a rubric (`proof-of-work-scoring`). ## Configuration If `.calibratr/config.json` exists, use `company.values` / `target_signals` / `exclusion_signals` as the Company DNA when judging fit, and `sourcing.use_paid_providers` to decide whether to spend credits on paid contact lookups. See the plugin's CONFIGURATION.md. ## Workflow 1. **Resolve identity.** From a resume/name/URL, extract: full name, primary email, and any **verbatim** handles/URLs (github_username, linkedin_url, twitter, portfolio). Never construct a handle — if it's not stated, leave it null. (See the verbatim-URL rule