Pipeline & Project Health: Project Candidate Completeness Audit

Prompt: Project Candidate Completeness Audit

Description: For a single project, audits every in-consideration candidate against a data-completeness checklist so the team knows which records need work before the next client update.
Frequency: Ad hoc (before client updates, slate presentations, or weekly project reviews)
Connectors: Clockwork Recruiting

Prompt: (copy/paste the prompt below into your chat in your LLM)

Project: [project name or ID]. Build a Candidate Completeness Audit using Clockwork Recruiting. Fetch the project with ? include=client_company,project_type and display as <Company> \ <Project> throughout.

Step 1 — Confirm scope. Fetch statuses via list_reference_data(data_type=statuses). Define "in consideration" as any candidacy with status.rank >= 1100. Confirm the boundaries with me if ambiguous.

Step 2 — Pull in-scope candidacies via list_candidacies with include=person,status,overview_note,assessment_note,next_steps_note. Paginate fully.

Step 3 — For each candidate, pull get_person_details for positions, compensation, education, emails, phone_numbers, tags, and notes (include=author).

Step 4 — Score each candidate against this checklist. Mark each ✅ present, ⚠️partial, or ❌ missing:

Current position (title + company + start date)
Work history (3+ prior positions, or full career if shorter)
Education (at least one entry)
Compensation (current comp — base, bonus/variable, equity if applicable)
Contact info (at least one email AND one phone number)
Résumé / CV attached — verify via get_person_details(detail_type=attachments). The response includes attachment_type, file_file_name, and a signed download_url (time-limited — use immediately, don't cache).
Overview note on the candidacy (present, non-empty)
Assessment note on the candidacy (present, non-empty)
Next steps note (present, dated within last 14 days)
Recent activity (any candidacy note in last 14 days)
Tags (at least one on the person record)

Step 5 — Compute a completeness score per candidate (count of ✅ out of 11, as %).

Output:

Header — <Company> \ <Project>, total in-scope candidates, average completeness score, audit date.
Candidate Completeness Table — one row per candidate, a column per checklist item plus the score, sorted by score ascending (worst first).
Top Gaps — 3–5 bullets on the most common missing fields ("6 of 9 candidates are missing compensation data").
Action List — prioritized list of candidates and fields to fix, grouped by likely owner (most recent note author). Format: [Candidate] — [missing fields] — suggested owner: [user].
Client-Ready Check — a GO / NOT YET verdict on whether the slate is ready to share, with reasoning.

Adapt It: Add screening-question ratings (get_candidacy(include_ratings=true)) as a completeness dimension Loosen scope to the whole pipeline for a data-hygiene sweep Tailor the checklist to your firm's must-haves Run it across all active searches and rank projects by average completeness.1.