--- name: caio-candidate-screener description: > Evaluates a prospective Chief AI Officer (CAIO) candidate and produces a rich, interactive HTML screening report with fit scores, strengths, risks, and a hire recommendation. Use this skill whenever a user pastes in a LinkedIn profile, CV, resume, or cover letter and wants to know how well the candidate fits a CAIO role. Also triggers for phrases like "screen this candidate", "how good is this person for the CAIO role", "rate this CV", "assess this LinkedIn profile", "is this person a good fit", "evaluate this applicant", or "score this hire". Always use this skill — even for informal inputs like a copied LinkedIn About section or a rough paste of work history — to generate the full structured report rather than answering in chat. --- # CAIO Candidate Screener Turn any candidate input — LinkedIn profile, CV, cover letter, or pasted bio — into a complete, interactive HTML screening report for a Chief AI Officer role. ## What this skill produces A single-file HTML report with 6 navigable tabs: 1. **Candidate Snapshot** — name, current role, location, years of experience, input type detected 2. **Fit Scores** — 7 scored dimensions with visual bars and a headline Overall Fit score 3. **Strengths** — what makes this candidate compelling for the CAIO role (specific, evidence-based) 4. **Red Flags & Gaps** — honest risks, missing experience, or misalignment signals 5. **Interview Angles** — 5 tailored questions to probe gaps or validate strengths 6. **Hire Recommendation** — Strong Yes / Yes / Maybe / No with a concise rationale paragraph --- ## Brand configuration Default palette (replace with your own brand colours as needed): ``` --yellow: #f3af00 --blue: #207796 --light-blue: #dff3fa --charcoal: #201600 --white: #ffffff Font: Muli (Google Fonts) ``` --- ## The CAIO role context (customise for your company) Before running the screener, define the role context. Default assumptions (edit for your company): - **Company stage**: Seed-stage / scale-up, founder-led - **Reporting line**: Direct to CEO - **Core mandate**: Deploy AI across the business; build AI culture; drive commercial AI outcomes - **Ideal candidate profile**: - 8+ years experience, 3+ in AI leadership - Relevant domain exposure (adapt to your industry) - Comfortable at a fast-moving, founder-led company - Can translate AI concepts to non-technical executives - Educator and culture-builder, not just a technologist - Commercially aware — ties AI to revenue and efficiency outcomes **Customise this section in your copy of the skill to match your company, stage, and industry.** --- ## Scoring dimensions (0–100) Evaluate each dimension based on evidence in the candidate input. If evidence is absent, score conservatively (don't assume). | Dimension | What to look for | |---|---| | **AI Leadership Experience** | Have they led AI initiatives at a meaningful scale? Senior title alone is not enough — look for shipped outcomes. | | **Domain / Industry Fit** | Any exposure to your industry or a closely related sector? | | **Business Translation Ability** | Do they write/talk about AI in business terms, not just tech terms? Do they reference ROI, efficiency, adoption? | | **Culture & People Development** | Evidence of building teams, running training, mentoring, or creating AI adoption programmes? | | **Commercial / Revenue Mindset** | Did they connect AI work to revenue, cost savings, client outcomes, or growth? | | **Startup / Scale-up Fit** | Have they worked in founder-led, fast-moving, resource-constrained environments? Big-company only = lower score. | | **Executive Presence** | Does the profile/CV communicate clearly and confidently? Would this person hold a room with your clients? | Score colour coding: green ≥65 · orange 40–64 · red <40 **Overall Fit** = weighted average (AI Leadership 25%, Domain 20%, Business Translation 15%, Culture 15%, Commercial 10%, Startup Fit 10%, Executive Presence 5%) --- ## Analysis process Before writing any HTML, work through the candidate input methodically: ### 1. Detect input type Identify whether the input is a LinkedIn profile, CV/resume, cover letter, or mixed. Note any gaps (e.g. "LinkedIn only — no detail on day-to-day responsibilities"). ### 2. Extract key facts - Full name, current title, current company, location - Total years of experience - AI-specific roles/projects (with years) - Industry exposure (especially to your target sector) - Team size managed - Educational background - Notable achievements (with metrics if stated) ### 3. Score each dimension For each of the 7 dimensions: assign a score 0–100, write 1–2 sentences of evidence-based justification. Be honest. A polished LinkedIn profile can mask weak AI depth — look for substance over labels. ### 4. Identify strengths (max 5) Each strength must be: - Specific to this candidate (not generic) - Tied to a concrete piece of evidence from the input - Framed in the context of what it means for the CAIO role ### 5. Identify red flags / gaps (max 5) Each flag must be: - Honest and specific — not hypothetical - Differentiated: "No industry exposure" is a gap; "AI buzzwords without shipped projects" is a red flag - Rated: Minor / Moderate / Significant ### 6. Generate interview questions (exactly 5) Tailor questions to this specific candidate — probe their weakest scoring areas while leaving room for them to shine on strengths. Format each question with a brief "(Why ask this)" note. ### 7. Hire recommendation Choose one: **Strong Yes** / **Yes** / **Maybe** / **No** Write 3–4 sentences of rationale grounded in the scores and evidence. Be direct — the hiring manager needs a clear signal, not a hedge. --- ## Output: HTML report Build one self-contained HTML file. The file must: - Use your brand colours and a single Google Font - Be fully self-contained — no external JS except Google Fonts - Work on mobile (responsive, horizontal-scroll tabs) - Show score bars that animate on load (CSS transitions) - Have expandable cards for red flags (click to reveal severity + detail) - Have a prominent, colour-coded recommendation badge at top of Hire Recommendation tab --- ## Delivering the output Save the HTML to `/mnt/user-data/outputs/_CAIO_screening.html`. Use `present_files` to share it with the user. Follow with a 2-sentence plain-language verdict: overall fit score and the single most important factor driving the recommendation.