Build a faith companion that lives in iMessage
Flint is a Christian companion people text for prayer, verses and daily check-ins, with no app to download.
Prepare and rehearse a salary negotiation
An employee used ChatGPT to organize market data and their results, rehearse objections, and turn a 4% offer into an 11% raise.
- Difficulty
- BeginnerOur rating
- Time to build
- Not reportedOur estimate
- Monthly cost
- Not reported
- Revenue
- Not reportedWe only show revenue the source states
- Engagement
- No public countThe source shows no number
What they did
They gave ChatGPT market data, performance reviews and project outcomes, then practiced their manager’s likely pushback.
How it went
They countered a 4% offer and settled at 11%, which they put at about $14,000.
Worth knowing
Use real market data; do not let it invent benchmarks.
Builder recipe
Our own starting point, written from the source. Adapt it to your setup.
Stack
- ChatGPT
Steps
- Give the agent access only to the accounts it needs.
- Write a clear brief with hard constraints.
- Ask it to report back before anything irreversible.
- Review the outcome and refine the brief.
Starter prompt
Here are my reviews, wins and market salary data. Build my case in 5 bullets, then play my manager and push back hard so I can practice.Tags
Read the original
Related use cases
Let agents find US doctors with checkable evidence over MCP
A developer built a remote MCP server with 1.2M US physicians, Medicare procedure volumes and facts from hospital sites.
Have your agent phone a restaurant and book only if your must-haves are met
A user’s OpenClaw agent called restaurants in Dubai with a strict brief and declined to book when a must-have was missing.
Build an agent that files refunds and insurance claims for you
A hackathon team built Recourse, which organizes evidence, prepares paperwork and calls customer support for you.
The Daily 5.
Five real AI agent use cases each morning, each with a recipe and a source. Plus a Sunday roundup. Free.