> ## Documentation Index
> Fetch the complete documentation index at: https://keystroke.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Use cases
> Common ways teams use Keystroke.
export const UseCase = ({title, prompt, apps = [], children}) => {
const [copied, setCopied] = useState(false);
const handleCopy = () => {
navigator.clipboard.writeText(prompt).then(() => {
setCopied(true);
setTimeout(() => setCopied(false), 2000);
}).catch(err => console.error("Failed to copy prompt:", err));
};
return ;
};
Teams use Keystroke to build internal agents and AI systems across every function. You can click any use case listed below to copy a ready-to-use prompt — paste it into the [platform agent](/docs/build-with-ai/platform-agent) chat in the web app, or hand it to your coding agent.
## For everyone
Manages your inbox and calendar, prepares you for meetings, and keeps every follow-up from slipping.
When a call ends, turns the transcript into action items, opens follow-up tasks, and sends everyone a crisp recap.
Gives a persistent company agent the context and skills to work across your team's systems.
A persistent AI coworker for leadership that can investigate, report, and remember useful context.
Researches a market and delivers a sourced, structured report instead of a chat response.
Keeps a running list of promises you made in email and Slack so nothing quietly slips.
## Sales
When a new meeting is booked, researches the attendee and company, then posts a concise prep note.
Researches target accounts, maintains CRM context, and drafts personalized outreach without sending.
When a prospect replies, turns the thread and CRM context into a thoughtful response for review.
Combines product, billing, and CRM signals to decide which accounts deserve attention now.
Turns agreed pricing from a customer call into a reviewable quote and follow-up package.
Keeps qualification, risks, and next steps current from every customer conversation.
## Marketing
Coordinates specialist agents for positioning, content, SEO, social, and email with shared context.
Every week, turns community signals and web research into topic clusters, article briefs, and Notion drafts.
Tracks meaningful competitor changes and explains what they mean for your positioning.
Turns product work and market signals into reviewable social and editorial drafts on a schedule.
Tests frontend changes and keeps the corresponding support documentation current.
Updates positioning docs with the words customers actually use on calls.
## Data
Documents your data, maintains shared meaning, answers questions, and creates dashboards on demand.
Pairs a conversational analyst with a model engineer that can improve the data layer through reviewed code.
When you ship a release, pulls adoption and retention for the launch cohort and posts a short recap.
Every week, flags at-risk accounts from product usage, support volume, and CRM signals.
Detects important metric shifts, investigates likely causes, and preserves the analysis for the team.
Every week, turns support conversations into themes with evidence and shares them with the team.
## Finance
Turns pricing agreed on calls into accurate, approval-gated quotes and customer drafts.
Learns from prior bookkeeping and asks for help only when a transaction is unclear.
Every Friday, matches payouts against your ledger, flags variances over threshold, and posts the report to finance.
Matches messy vendor invoices against records and routes only meaningful exceptions for approval.
A few times a week, finds overdue invoices and drafts collection emails for review.
Every month, reviews recurring Ramp charges and related receipts to flag spend worth a second look.
## Operations
Investigates customer bugs across support history, documentation, and code before preparing the response.
When a new hire starts, sends a first-week checklist, schedules intros, and answers questions from company docs.
When a contract is received, flags unusual terms, tracks deadlines, and prepares a summary for attorney review.
Routes every meeting action to the right teammate before it gets forgotten.
Turns recurring support questions into draft help articles for human review.
Keeps a running record of what leadership decided, why, and who owns the follow-through.
## Engineering
Investigates the most severe and frequent production issue and carries its fix through a reviewable PR.
Finds the error hurting the most customers, traces it through the code, and prepares the fix.
Finds the database query with the greatest performance cost and opens a code-level fix.
When a PR opens, scores blast radius, requests reviewers, and comments with a risk summary.
Lets support summon an engineer that investigates, fixes, and closes the loop with the customer.
After a production deploy, checks errors and metrics and posts a go / no-go to the release channel.