30 Claude Prompts That Build Support Dashboards

You are a support operations analyst who builds resolution SLA boards, not first-reply clocks. <context> I need a resolution SLA dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic solved tickets so I can click the working screen in the chat. </context> <inputs> - Resolution target: [RESOLUTION TARGET] - Channel overrides: [E.G. CHAT 4H, EMAIL 24H, PHONE SAME SHIFT] - Priority overrides: [E.G. P1 4H, P2 24H] - Date range: [E.G. LAST 30 DAYS] - Team: [NAME] </inputs> <task> Build the dashboard with: KPI cards (median time to resolve, percent inside [RESOLUTION TARGET], misses, tickets still open past target); a line of daily resolution hit rate; a bar by channel and by priority; a distribution of hours-to-solve; and a miss log (ticket, channel, priority, hours, target, owner). Compute hours-to-solve from opened-at and solved-at. Open tickets past target sit in a separate still-open list. Clock rule labeled on the page. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Tickets and targets in labeled JS constants at the top. - Misses flagged red. This board is resolution, not first response. Accessible contrast. - State calendar time versus business hours. Do not hide that choice. </constraints> <format> Return the full HTML file, then explain time-to-resolve math and which array to replace with a solved-ticket export. </format>

You are a support operations analyst who builds SLA matrix boards that a weekly review can argue from. <context> I need an SLA-by-priority-and-plan dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic tickets so I can click the working screen in the chat. </context> <inputs> - SLA matrix: [SLA MATRIX] - Priorities: [E.G. P1-P4] - Plans: [E.G. ENTERPRISE, PRO, FREE] - Metric: [FIRST REPLY / NEXT REPLY / RESOLUTION] - Date range: [E.G. LAST 30 DAYS] </inputs> <task> Build the dashboard with: a metric toggle; KPI cards (cells on track, cells missing, worst cell, tickets in the worst cell); a priority-by-plan heatmap of hit rate versus that cell's target from [SLA MATRIX]; a click-through list of tickets in the selected cell; a trend for the selected cell; and a matrix table that shows target, n, hit rate, and median time. Compute hit/miss from each ticket's priority, plan, and the matching target. Small-n cells marked when count is below an editable threshold. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Tickets and the matrix in labeled JS constants at the top. - Heatmap not color-only (show the percent). Accessible contrast. - Do not invent a target I left blank. Mark [VERIFY]. </constraints> <format> Return the full HTML file, then explain cell matching and which constants to replace. </format>

You are a support operations analyst who builds next-reply clocks that catch a silent thread before it breaches. <context> I need a next-reply SLA dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic waiting-on-us tickets so I can click the working screen in the chat. </context> <inputs> - Next-reply target: [NEXT REPLY TARGET] - Channel overrides: [E.G. CHAT 15M, EMAIL 8H] - Waiting-on-us statuses: [E.G. OPEN, PENDING-AGENT] - Team: [NAME] - Business hours: [E.G. 9-17 WEEKDAYS] or 24/7 </inputs> <task> Build the dashboard with: KPI cards (waiting on us, median minutes since last customer message, percent inside [NEXT REPLY TARGET], tickets inside 20% of the clock); a ticking watchlist sorted soonest-to-breach with minutes remaining; a channel bar; a closed-thread table of actual next-reply versus target; and a paused-outside-hours note if I asked for business hours. Compute remaining time from last-customer-at, channel target, and a now constant. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Tickets and SLA constants at the top in labeled JS. - Under 20% remaining amber; breached red. Accessible contrast. - This is next reply, not first response. Show that on the header. </constraints> <format> Return the full HTML file, then explain remaining-time math and how to paste a waiting-on-us export. </format>

You are a support operations analyst who builds SLA boards that pause outside the hours you actually staff. <context> I need a business-hours SLA dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic ticket timestamps so I can click the working screen in the chat. </context> <inputs> - Business hours: [BUSINESS HOURS] - Timezone: [E.G. AMERICA/NEW_YORK] - Closed days: [E.G. SAT, SUN, LISTED HOLIDAYS] - SLA metric: [FIRST REPLY / RESOLUTION] - Target inside those hours: [E.G. 8 BUSINESS HOURS] </inputs> <task> Build the dashboard with: a header that prints [BUSINESS HOURS] and the timezone; KPI cards (calendar-hour median, business-hour median, hit rate on business clock, tickets that look late only on calendar time); a side-by-side bar of calendar versus business hit rate; a holiday list from constants; a ticket table with both clocks; and a toggle that recounts hit/miss on calendar time so I can see the lie. Compute business elapsed time in JS by skipping closed intervals. Show the skip rule on the page. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Hours, holidays, and tickets in labeled JS constants at the top. - Timezone labeled. Accessible contrast. - Do not silently assume 24/7 if I gave hours. </constraints> <format> Return the full HTML file, then write the business-elapsed function in a short note and which constants to replace. </format>

You are a support operations analyst who builds breach-credit boards that sit next to the plan you sold, not in a separate myth. <context> I need an SLA breach credit dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic breach rows so I can click the working screen in the chat. </context> <inputs> - Credit rule: [CREDIT RULE] - Plans that earn credits: [E.G. ENTERPRISE, PRO] - SLA that triggers credit: [E.G. P1 RESOLUTION] - Date range: [E.G. THIS MONTH] - Currency: [E.G. USD] </inputs> <task> Build the dashboard with: KPI cards (breaches, accounts hit, credits estimated, credits already issued if I gave that field); a bar of breaches by plan; an account table (account, plan, breaches, credit estimate, status); a miss log; and a one-line rule note I can edit. Compute credit estimates from [CREDIT RULE] and sample breach rows. Mark estimates as estimates, not invoices. Clicking an account filters the miss log. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Breaches and the rule in labeled JS constants at the top. - Currency labeled. Accessible contrast. - Do not invent legal language. Arithmetic from my rule only. </constraints> <format> Return the full HTML file, then write the credit formula and which arrays to replace. </format>

You are a support quality analyst who builds CSAT trend boards a weekly review can trust. <context> I need a CSAT trend dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic survey rows so I can click the working screen in the chat. </context> <inputs> - Survey source: [SURVEY SOURCE] - Score scale: [E.G. 1-5, CSAT = 4 AND 5] - Date range: [E.G. LAST 12 WEEKS] - Queues: [E.G. BILLING, TECH] - Channels: [E.G. EMAIL, CHAT] </inputs> <task> Build the dashboard with: KPI cards (CSAT %, average score, surveys scored, response rate); a weekly line of CSAT % with a response-rate area behind it; a queue and channel split; a week table (sent, scored, CSAT %, average); and a click on a week that lists those surveys. Compute CSAT as scores meeting the 4-and-5 rule (or my rule) divided by scored surveys. Response rate = scored / sent. Small-n weeks marked. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Survey rows and send counts in labeled JS constants at the top. - Show the CSAT rule on the page. Accessible contrast. - Do not invent NPS unless I add it later. </constraints> <format> Return the full HTML file, then write CSAT and response-rate formulas and which arrays to replace. </format>

You are a support quality analyst who builds detractor watchlists that turn a low score into a next action. <context> I need a CSAT detractor dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic low-score surveys so I can click the working screen in the chat. </context> <inputs> - Detractor threshold: [CSAT THRESHOLD] - Fields: [SCORE, VERBATIM, TICKET, TOPIC, AGENT, ACCOUNT, DATE] - Follow-up statuses: [E.G. UNTOUCHED, EMAILED, CLOSED LOOP] - Date range: [E.G. LAST 30 DAYS] - Owners: [NAMES OR TEAMS] </inputs> <task> Build the dashboard with: KPI cards (detractors, untouched, closed-loop rate, repeat detractor accounts); a topic bar of low scores; a watchlist table sorted newest first with verbatim, ticket, owner, and a status control; a filter for untouched; and a closed-loop note that a status change updates the KPI without a reload. Detractor = score at or below [CSAT THRESHOLD]. Status is a dropdown I can click. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Surveys in labeled JS constants at the top. - Verbatim visible, truncated with expand. Accessible contrast. - Do not invent angry quotes I did not seed. Sample verbatim must sound like a customer, not lorem. </constraints> <format> Return the full HTML file, then explain the threshold and which array to replace with a survey export. </format>

You are a support quality analyst who builds CSAT boards that are fair to agents and honest about sample size. <context> I need a CSAT-by-queue-and-agent dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic survey rows so I can click the working screen in the chat. </context> <inputs> - Team: [TEAM] - Agents: [NAMES] - Queues: [LIST] - Small-n rule: [E.G. UNDER 15 SCORED] - Date range: [E.G. LAST 30 DAYS] - CSAT rule: [E.G. 4 AND 5 ON A 1-5] </inputs> <task> Build the dashboard with: a queue filter; KPI cards (team CSAT, agents above team, agents below, agents in small-n); a bar of CSAT by queue with n labels; an agent table (scored, CSAT %, average, versus team) that marks small-n instead of ranking them; a toggle to include or hide small-n; and a selected-agent sparkline. Compute every rate from survey rows. Do not rank small-n agents as best or worst. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Surveys in labeled JS constants at the top. - Small-n labeled in text, not color alone. Accessible contrast. - This is quality, not a public leaderboard. No hero or slacker labels. </constraints> <format> Return the full HTML file, then write the small-n rule and which array to replace. </format>

You are a support quality analyst who builds survey-funnel boards so a quiet CSAT is not mistaken for a happy book. <context> I need a survey response-rate dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic send and reply counts so I can click the working screen in the chat. </context> <inputs> - Survey window: [SURVEY WINDOW] - Steps: [SENT, OPENED, STARTED, SCORED] or my list - Split by: [CHANNEL / QUEUE / PLAN] - Target response rate: [E.G. 20%] - Date range: [E.G. LAST 8 WEEKS] </inputs> <task> Build the dashboard with: KPI cards (sent, scored, response rate, versus [TARGET RESPONSE RATE]); a funnel of the steps; a grouped bar of response rate by the split I gave; a weekly line of sent versus scored; and a table of the split with each step and drop-off. Compute each rate from step counts. Clicking a split row filters the weekly line. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Step counts in labeled JS constants at the top. - Below-target splits flagged. Accessible contrast. - If I have no opened count, mark that step [VERIFY] and skip it, do not invent opens. </constraints> <format> Return the full HTML file, then write funnel math and which constants to replace. </format>

You are a support quality analyst who builds reason-level CSAT boards that tell product which pain is actually scoring low. <context> I need a CSAT-versus-contact-reason dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic tagged surveys so I can click the working screen in the chat. </context> <inputs> - Reason codes: [REASON CODES] - CSAT rule: [E.G. 4 AND 5 ON A 1-5] - Date range: [E.G. LAST 60 DAYS] - Small-n rule: [E.G. UNDER 10] - Queues: [OPTIONAL] </inputs> <task> Build the dashboard with: KPI cards (lowest-scoring reason with enough n, highest-volume reason CSAT, reasons below team, untagged share); a scatter or grouped view of volume versus CSAT by reason; a ranked pain list (reason, n, CSAT %, versus team); a click that lists those surveys and verbatims; and an untagged bucket. Compute CSAT per reason from tagged rows. Small-n reasons sit in a separate muted list. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Surveys with reason tags in labeled JS constants at the top. - Do not crown a small-n reason as worst. Accessible contrast. - Do not invent product bug ids. </constraints> <format> Return the full HTML file, then explain grouping and which array to replace with tagged surveys. </format>

You are a support operations analyst who builds backlog burndowns a founder can use to decide when to hire or freeze intake. <context> I need a backlog burndown dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic weekly open counts so I can click the working screen in the chat. </context> <inputs> - Backlog scope: [BACKLOG SCOPE] - History: [E.G. LAST 8 WEEKS OF OPEN, ARRIVALS, SOLVES] - Starting open: [COUNT] - Target open: [E.G. 80] or none - Productive solve capacity now: [E.G. 220 / WEEK] </inputs> <task> Build the dashboard with: KPI cards (open now, weekly net, weeks to target at current net, weeks to target at stated capacity); a burndown of open versus a straight required-pace line if I gave a target; a bar of arrivals versus solves; a table of week, open, arrivals, solves, net; and a callout of the first week the pile grows again if arrivals stay at the recent median. Compute open from starting open plus weekly nets. Label the projection as a flat-rate estimate. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Weekly series in labeled JS constants at the top. - Label estimate versus measured. Accessible contrast. - Do not invent a hiring plan. Show the week and the gap. </constraints> <format> Return the full HTML file, then write burndown and weeks-to-clear math and which constants to replace. </format>

You are a support operations analyst who builds pending boards that split work we owe from work the customer owes. <context> I need a waiting-on-us versus waiting-on-customer dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic pending tickets so I can click the working screen in the chat. </context> <inputs> - Pending statuses: [PENDING STATUSES] - Age rule: [WAITING ON US FROM LAST CUSTOMER MSG; WAITING ON CUSTOMER FROM LAST AGENT MSG] - Age buckets: [E.G. 0-1D, 1-3D, 3-7D, 7D+] - Owners: [NAMES OR TEAMS] - Queues: [LIST] </inputs> <task> Build the dashboard with: status chips from [PENDING STATUSES]; KPI cards (waiting on us, waiting on customer, 7d+ on us, 7d+ on customer); a stacked bar of each side by age bucket; an owner table; a chase list for waiting-on-us sorted oldest last-customer-message first; and a nudge list for waiting-on-customer. Show the age rule on the page. Chips filter every widget. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Tickets in labeled JS constants at the top. - Waiting-on-us rows visually louder. Accessible contrast. - Do not mix the two clocks. If a timestamp is missing, mark [VERIFY]. </constraints> <format> Return the full HTML file, then explain both age rules and which array to replace. </format>

You are a support operations analyst who builds SLA-risk backlogs that a lead can empty in the last hour of the shift. <context> I need an SLA-risk backlog dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic open tickets so I can click the working screen in the chat. </context> <inputs> - Breach window: [BREACH WINDOW] - Risk band: [E.G. LAST 20% OF THE CLOCK] - SLA metric: [FIRST REPLY / NEXT REPLY / RESOLUTION] - Queues and channels: [LIST] - Staff on now: [COUNT OR NAMES] </inputs> <task> Build the dashboard with: KPI cards (open in scope, already breached, inside the risk band, minutes to the next breach); a histogram of minutes remaining; a queue split; a sorted risk table (ticket, queue, channel, metric, minutes remaining, assignee); and a filter for breached / risk / safe. Compute remaining time from due-at versus a now constant. Default sort is soonest to breach. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Tickets and due-ats in labeled JS constants at the top. - Breached red, risk amber. Filters must hide rows. Accessible contrast. - Do not invent paging. This is a list a lead can assign by hand. </constraints> <format> Return the full HTML file, then explain remaining-time math and which array to replace. </format>

You are a support operations analyst who builds reopen piles that show failed first closes still sitting in the book. <context> I need a reopened-ticket backlog dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic reopen events so I can click the working screen in the chat. </context> <inputs> - Reopen window: [REOPEN WINDOW] - Reason codes: [E.G. WRONG ANSWER, PRODUCT BUG, CAME BACK] - Statuses that count as still open: [E.G. REOPENED, OPEN] - Teams: [L1, L2, BILLING] - Date range for the rate: [E.G. LAST 30 DAYS] </inputs> <task> Build the dashboard with: KPI cards (open reopened tickets, reopen rate in the window, twice-plus count, median days since first solve); a reason bar; a bounce table (ticket, first solve, reopen count, reason, current owner, age since reopen); a twice-plus watchlist; and a weekly reopen-rate line. Reopen means solved then opened again inside [REOPEN WINDOW]. Show that rule. A reason chip filters the table. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Events in labeled JS constants at the top. - Twice-plus rows flagged. Accessible contrast. - This is a backlog of open reopens, not an escalation funnel. </constraints> <format> Return the full HTML file, then explain the reopen rule and which event array to replace. </format>

You are a support operations analyst who builds after-hours backlog boards that show what landed while nobody was on the clock. <context> I need an after-hours backlog growth dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic create timestamps so I can click the working screen in the chat. </context> <inputs> - Off hours: [OFF HOURS] - Timezone: [E.G. AMERICA/NEW_YORK] - Weekend definition: [E.G. FRI 18:00 TO MON 09:00] - Date range: [E.G. LAST 8 WEEKS] - Queues: [LIST] </inputs> <task> Build the dashboard with: KPI cards (off-hours arrivals, share of all arrivals, Monday-open leftover, median Monday age); a stacked weekly bar of weekday-night versus weekend arrivals; a Monday-morning leftover line; a queue split; and a leftover table of tickets still open that were created off hours. Compute off-hours flags from create timestamps and [OFF HOURS]. Timezone labeled. Week chips filter the leftover table. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Tickets in labeled JS constants at the top. - Timezone and weekend rule on the page. Accessible contrast. - Do not invent on-call pages. Arrivals and leftovers only. </constraints> <format> Return the full HTML file, then write the off-hours flag rule and which array to replace. </format>

You are a support insights analyst who builds topic-mix boards that a weekly triage can act on. <context> I need a topic volume mix dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic tagged tickets so I can click the working screen in the chat. </context> <inputs> - Topic tags: [TOPIC TAGS] - Date range: [E.G. LAST 8 WEEKS] - Rising rule: [E.G. SHARE UP 50% WEEK OVER WEEK] - Untagged handling: [SHOW AS UNTAGGED] - Queues: [OPTIONAL] </inputs> <task> Build the dashboard with: KPI cards (tagged tickets, untagged share, top tag, rising tags); a stacked weekly area by tag; a period donut; a rising-tag list using my rule; a tag table (volume, share, wow change, CSAT if present); and a click on a tag that lists sample subjects. Compute share and wow from weekly tag counts. Keep an untagged bucket. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Ticket-tag rows in labeled JS constants at the top. - Rising rule on the page. Accessible contrast. - Do not invent a tag I did not list except untagged. </constraints> <format> Return the full HTML file, then write share and rising-rule math and which array to replace. </format>

You are a support insights analyst who builds emerging-topic boards that catch a new failure before it has a tidy tag. <context> I need an emerging-topic dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic ticket subjects so I can click the working screen in the chat. </context> <inputs> - Lookback: [LOOKBACK] - Cluster key: [E.G. NORMALIZED SUBJECT PHRASE] - New rule: [E.G. FIRST SEEN IN THE LAST 7 DAYS AND 5+ TICKETS] - Product areas: [LIST] - Sample subjects: [PASTE OR ASK CLAUDE TO INVENT 40] </inputs> <task> Build the dashboard with: KPI cards (new clusters, tickets in those clusters, clusters with no owner, clusters with no existing tag); a cluster table (phrase, count, first seen, last seen, area, owner, status); a 14-day strip for the selected cluster; a no-owner watchlist; and a simple JS normalizer (lowercase, strip ids and punctuation) documented on the page. Selecting a cluster lists member subjects. Compute first seen from sample created-ats. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Subjects in labeled JS constants at the top. - Show the new rule. Accessible contrast. - This is a phrase cluster, not a machine-learning claim. Label it as simple grouping. </constraints> <format> Return the full HTML file, then explain the normalizer, the new rule, and which array to replace. </format>

You are a support insights analyst who builds macro boards that show which canned answers are leaking reopens. <context> I need a macro hit-rate dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic macro-send rows so I can click the working screen in the chat. </context> <inputs> - Macro list: [MACRO LIST] - Join fields: [TICKET, MACRO, SENT AT, REOPENED IN 7D, CSAT] - Weak rule: [E.G. REOPEN AFTER SEND ABOVE 15% OR CSAT BELOW TEAM] - Date range: [E.G. LAST 60 DAYS] - Queues: [OPTIONAL] </inputs> <task> Build the dashboard with: KPI cards (macros tracked, sends, reopen-after-send rate, macros that fail the weak rule); a scatter of use count versus reopen rate; a table (macro, sends, reopen %, CSAT %, weak flag); a click that lists tickets that reopened after that macro; and a unused-macro list if I included macros with zero sends. Compute rates from the join. Weak flags from my rule. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Send rows and the macro list in labeled JS constants at the top. - Show the weak rule. Accessible contrast. - Do not rewrite the macros. Numbers only. </constraints> <format> Return the full HTML file, then write reopen-after-send math and which arrays to replace. </format>

You are a support insights analyst who builds knowledge-gap boards that tell docs what to write next. <context> I need a knowledge-gap dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic tags and article rows so I can click the working screen in the chat. </context> <inputs> - Help center map: [HELP CENTER MAP] - Topic tags: [LIST] - Article fields: [SLUG, TITLE, TAGS, LAST UPDATED, STATUS] - Gap rule: [E.G. 20+ TICKETS IN 30 DAYS AND NO LIVE ARTICLE] - Stale rule: [E.G. ARTICLE OLDER THAN 180 DAYS] </inputs> <task> Build the dashboard with: KPI cards (gap topics, tickets in gaps, stale articles, topics with an article that still has high volume); a ranked gap list (tag, tickets, suggested title from the tag name, owner blank); a stale-article table; a matched view of tag versus article; and a filter for gap / stale / covered. Compute gaps by joining tag volume to articles that share that tag and are live. Show both rules on the page. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Tags and articles in labeled JS constants at the top. - Do not invent article bodies. Titles are suggestions I can edit. Accessible contrast. - Suggested titles are not published docs. </constraints> <format> Return the full HTML file, then explain the join and which arrays to replace with a help-center export. </format>

You are a support insights analyst who builds product-area maps that support and product can share in the same standup. <context> I need a product-area topic dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic product-tagged tickets so I can click the working screen in the chat. </context> <inputs> - Product areas: [PRODUCT AREAS] - Date range: [E.G. LAST 30 DAYS] - Metrics: [VOLUME, CSAT, REOPEN RATE, MEDIAN RESOLVE] - Handoff rule: [E.G. TAGGED BUG AND STILL OPEN 3+ DAYS] - Owners: [SQUADS OR PMS] </inputs> <task> Build the dashboard with: an area grid of cards (volume, CSAT, reopen %, owner); a selected-area topic breakdown; a handoff list using my rule; a comparison table across areas; and a click on a card that filters every list. Compute metrics from tickets that have one primary area field. If a ticket has two areas, use the primary field only and say so. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Tickets in labeled JS constants at the top. - Cards are buttons, not just color. Accessible contrast. - Do not invent Jira statuses. Handoff is a tagged list I can copy. </constraints> <format> Return the full HTML file, then explain primary-area grouping and which array to replace. </format>

You are a support workforce analyst who builds forecast-versus-seats boards a lead can staff from before the day starts. <context> I need a forecast-versus-seats dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic interval forecasts so I can click the working screen in the chat. </context> <inputs> - Interval: [INTERVAL] - Date or typical weekday: [E.G. NEXT TUESDAY] - Forecast arrivals per interval: [COUNTS OR ASK TO INVENT A CURVE] - Average handle time minutes: [E.G. 8] - Occupancy target: [E.G. 80%] - Seats rostered: [PER INTERVAL] </inputs> <task> Build the dashboard with: KPI cards (peak interval, seats short at peak, surplus hours, intervals in gap); a grouped bar of seats needed versus seats rostered by interval; a gap table (interval, forecast, AHT, seats needed, rostered, gap); and a click on an interval that lists who is on. Seats needed = (arrivals * AHT minutes / interval minutes) / occupancy target. Show that formula. Label it a simple estimate, not an Erlang model. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Forecast and roster in labeled JS constants at the top. - Gap cells red, surplus muted. Formula on the page. Accessible contrast. - Do not auto-assign people. Show the gap only. </constraints> <format> Return the full HTML file, then write the seats-needed formula and which constants to replace. </format>

You are a support workforce analyst who builds skill-coverage boards that show the queue you cannot actually staff. <context> I need a skill coverage dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with a realistic skill-tagged roster so I can click the working screen in the chat. </context> <inputs> - Skill set: [SKILL SET] - Queues that need those skills: [MAP] - People and skills: [ROSTER] - Seats needed by skill for the shift: [COUNTS] - Shift: [E.G. TODAY 9-17] </inputs> <task> Build the dashboard with: a header for the shift; KPI cards (skills in gap, worst gap, unused surplus skills, people out); a bar of needed versus qualified-and-on seats by skill; a people table (name, skills, on/off, assigned queue); a spare-people list whose skills do not match the gap; and skill chips that filter the people table. Qualified-and-on means the person is on the shift and has that skill. Hour or shift is a single block unless I add intervals later. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Roster and demand in labeled JS constants at the top. - Gap cells red. Accessible contrast. - Do not auto-assign. Show the gap and the spare list only. </constraints> <format> Return the full HTML file, then explain qualified-and-on and which arrays to replace with a roster export. </format>

You are a support workforce analyst who builds shrinkage boards that subtract the hours people will not be in the queue. <context> I need a shrinkage and PTO dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic out rows so I can click the working screen in the chat. </context> <inputs> - Shift: [SHIFT] - Horizon: [E.G. NEXT 14 DAYS] - Planned seats per day: [COUNTS] - Out types: [PTO, MEETING, TRAINING, BREAK] - Shrinkage target: [E.G. 30%] - People: [NAMES] </inputs> <task> Build the dashboard with: KPI cards (planned hours, out hours, net hours, days over the shrinkage target); a stacked daily bar of in-queue versus each out type; a people-by-day heatmap of net hours; a PTO calendar table; and a thin-day callout listing who is out. Net = planned minus out. Shrinkage = out / planned. Cells are buttons that list that day's out rows. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Planned seats and out rows in labeled JS constants at the top. - Color plus a percent so the heatmap is not color-only. Accessible contrast. - No ranking language. Hours only. </constraints> <format> Return the full HTML file, then write net and shrinkage formulas and which arrays to replace. </format>

You are a support workforce analyst who builds new-hire ramp boards that are fair to people still learning the queue. <context> I need a new-hire queue ramp dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with realistic ramping agents so I can click the working screen in the chat. </context> <inputs> - Ramp plan: [RAMP PLAN] - Ramping agents and start dates: [NAMES AND DATES] - Full-target AHT, CSAT, and tickets per week at 100%: [NUMBERS] - Weeks in the plan: [E.G. 8] - Queue they sit: [NAME] </inputs> <task> Build the dashboard with: KPI cards (hires on ramp, on or ahead of plan, behind, average % of expected tickets); a line of expected versus actual tickets by ramp week; per-hire cards with start date, week number, expected AHT, actual AHT, expected CSAT, actual CSAT, and a behind flag; a table of the same; and a flag on anyone two bands behind. Compute expected values from start date and [RAMP PLAN]. Small-n CSAT marked. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Hires and weekly stats in labeled JS constants at the top. - Behind-plan cards flagged. Accessible contrast. - No slacker labels. Numbers versus the plan only. </constraints> <format> Return the full HTML file, then explain how ramp week is derived from start date and how to add a hire. </format>

You are a support workforce analyst who builds after-hours staffing boards that show the hour you will miss before it happens. <context> I need an after-hours staffing gap dashboard as one self-contained HTML file with inline CSS, vanilla JS, and a CDN chart library, seeded with a realistic night roster so I can click the working screen in the chat. </context> <inputs> - On-call window: [ON CALL WINDOW] - Timezone: [E.G. AMERICA/NEW_YORK] - Skills needed overnight: [E.G. TECH, BILLING] - Demand by hour: [SEATS OR ARRIVALS] or ask Claude to invent a light overnight curve - People and windows: [ROSTER] </inputs> <task> Build the dashboard with: a day picker inside a sample week; KPI cards (hours with a gap, worst gap, leftover tickets into morning, people on); a stacked bar of filled versus needed seats by hour and skill; a gap table (hour, skill, needed, filled, who is on); a leftover note for tickets still open at window end; and hour cells that click through to names. Compute filled from roster intervals overlapping that hour. Timezone labeled. </task> <constraints> - One self-contained HTML file; one CDN charting library only. - Roster and demand in labeled JS constants at the top. - Gap cells red. Accessible contrast. - Do not auto-assign. Show the gap and the names only. </constraints> <format> Return the full HTML file, then explain overlap math and which arrays to replace with a night-roster export. </format>