Build internal real estate tools with Codex
A rental owner used Codex to make custom tools for offer pricing, property reports, and contract tracking.
Agent Uses · 1,321 real examples
Coding agents that build, review, test, migrate and ship software. Each entry shows what the person asked for, how it went, and a prompt to try it yourself.
The day's AI news plus 3 agent uses you can try, with ready-to-paste prompts. Free; unsubscribe anytime.
A rental owner used Codex to make custom tools for offer pricing, property reports, and contract tracking.
They built a Tangled MCP server so Claude Code could check branches and handle issues for repos in owner/repo form.
Devin can take product work, implement it, and hand back a pull request for human review.
A reviewer put Manus through eight programming chores and found it dependable overall, with a few weak spots in validation and proof of execution.
They had Hermes Agent assemble a local AI stack and connect Gbrain for chat-based knowledge-graph lookups.
A team handed Devin narrow fixes in their repo, then reviewed the tests and pull requests it produced.
Devin created a small Node.js Express service, tested its routes, and added error handling in about 4 minutes.
A developer had Devin build a backend token endpoint from docs, then cleaned up the pull request through a few review cycles.
Cognition used Devin to turn Slack and Linear requests into pull requests, then keep fixing review and CI issues until checks passed.
Devin handled app scaffolding, repo debugging, and deployment setup in demos and early user reports.
A software engineer had Replit Agent turn Wefunder listings into a searchable Leaflet map, then fixed bad API results by loading the data themselves.
A developer used Replit Agent to make a Hacker News scraper with storage, an API, and a front end, then deployed it.
A developer had Replit Agent assemble a reporting app with a database, image storage, analytics, and a public landing page.
A founder used Cursor to inspect a trip-planning app, surface missing pieces, sketch screens, and then build a feature with parallel agents.
A screen-reader user got Muse to create an iCloud skill, but password handling and accessibility issues made it hard to use.
They built a GitHub bot that sorts issues, reviews pull requests, answers with citations, and opens patch PRs for humans to check.
Google’s Sashiko scans Linux kernel submissions and flags bugs that human reviewers missed.
A desktop agent connected to a mainframe to read COBOL, submit jobs, and summarize failures from spool output.
A custom agent can set up a Cloudflare account, buy a domain, and ship an app with only brief approval steps.
They assembled a call-answering agent with live transcription, generation, and speech output, then moved it closer to the services to get about 400ms latency.
Store project context in git so a coding agent can reuse decisions and notes across sessions.
They used a LangGraph agent to test CUDA kernel tweaks, verify results, and profile candidates with Nsight.
Create a multi-step system that assigns issues, tests changes, and leaves the final merge to a person.
Microsoft used agents to replace the Copilot runtime’s TypeScript with Rust while keeping the implementation’s overall design the same.