# Real Agent Cases

> A directory of 142 real, source-verified AI agent use cases. Each entry has the task, outcome, agent(s), evidence type, any claimed result (attributed to its source, not independently verified), a beginner-friendly flag, a reusable prompt (labeled original or recreated), and the original source link and date.

- [About](https://www.realagentcases.com/about/)
- [How we verify](https://www.realagentcases.com/how-we-verify/)
- [All agents](https://www.realagentcases.com/agent/)
- [Sitemap](https://www.realagentcases.com/sitemap.xml)
- [Full version with every prompt](https://www.realagentcases.com/llms-full.txt)

## Coding

### Migrate two websites from WordPress to Wagtail with Claude Code
- URL: https://www.realagentcases.com/use-cases/migrate-two-websites-from-wordpress-to-wagtail-with-claude-code/
- Outcome: Over a weekend Claude Code migrated two sites from WordPress to Wagtail (PHP to Python), including server setup, data transfer, testing and fixes, and redesigned a homepage; the poster reports nothing broke and traffic and tools stayed in place.
- Agents: Claude Code
- Category: Coding
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/xamsor/status/2021513354431332677 (2026-02-11)
- Last verified: 2026-10-16

### Port a backend to Convex in four hours with Factory's Droid
- URL: https://www.realagentcases.com/use-cases/port-a-backend-to-convex-in-four-hours-with-factory-s-droid/
- Outcome: The poster reports that Factory's Droid ported their tRPC/Drizzle custom backend to Convex in about 4 hours, with some bugs left to fix.
- Agents: Factory Droid
- Category: Coding
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/Dallenpyrah/status/1971996302746824729 (2025-09-27)
- Last verified: 2026-10-16

### Migrate a 50-screen app from Firebase to Supabase with Claude Code
- URL: https://www.realagentcases.com/use-cases/migrate-a-50-screen-app-from-firebase-to-supabase-with-claude-code/
- Outcome: Going feature by feature, the poster used Claude Code to generate database schemas, write queries and fix daily errors; they report migrating about 85% of a 45–50 screen Flutter app from Firebase to Supabase in 10–20 days.
- Agents: Claude Code
- Category: Coding
- Evidence: Described by poster
- Claimed result: ~85% migrated in 10–20 days (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/prayag_dalal/status/1963555583392714791 (2025-09-04)
- Last verified: 2026-10-16

### Build a blog feature by voice with Codex while cooking
- URL: https://www.realagentcases.com/use-cases/build-a-blog-feature-by-voice-with-codex-while-cooking/
- Outcome: Talking to Codex in voice mode while cooking dinner, the author got a Django newsletters archive with four importers and a pull request, then finished the rest with typed prompts.
- Agents: OpenAI Codex (ChatGPT desktop, voice mode)
- Category: Coding
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://simonwillison.net/2026/Oct/9/built-using-my-voice/ (2026-10-09)
- Last verified: 2026-10-15

### Migrate 180 microservices to a new framework version with Cursor agents
- URL: https://www.realagentcases.com/use-cases/migrate-180-microservices-to-a-new-framework-version-with-cursor-agents/
- Outcome: Brex reports that its self-correcting, self-hosted Cursor agents let one engineer migrate 180 microservices from Micronaut 3 to 4 in a month, which it calls a 32x per-capita speedup.
- Agents: Cursor (self-hosted agents)
- Category: Coding
- Evidence: Company case study
- Claimed result: 180 microservices migrated in one month (According to Brex; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://www.brex.com/journal/agent-automations-for-big-migrations (2026-07-22)
- Last verified: 2026-10-15

### Automate defect remediation across C/C++ codebases with Codex CLI
- URL: https://www.realagentcases.com/use-cases/automate-defect-remediation-across-c-c-codebases-with-codex-cli/
- Outcome: According to OpenAI's case study, Cisco's Codex CLI defect-repair workflow raised defect resolution throughput 10–15x, and Codex build optimization saved more than 1,500 engineering hours a month.
- Agents: OpenAI Codex CLI
- Category: Coding
- Evidence: Company case study
- Claimed result: 10–15x defect resolution throughput (According to OpenAI; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://openai.com/index/cisco/ (2026-05-27)
- Last verified: 2026-10-15

### Port an HTML5 parser from Python to JavaScript with Codex CLI
- URL: https://www.realagentcases.com/use-cases/port-an-html5-parser-from-python-to-javascript-with-codex-cli/
- Outcome: Codex CLI (GPT-5.2) ran for hours largely unattended and produced about 9,000 lines of dependency-free JavaScript passing 9,200 html5lib tests, in about 4.5 hours start to finish.
- Agents: OpenAI Codex CLI
- Category: Coding
- Evidence: Screenshots or logs
- Claimed result: Passes 9,200 html5lib tests (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://simonwillison.net/2025/Dec/15/porting-justhtml/ (2025-12-15)
- Last verified: 2026-10-15

### Upgrade a plugin to a new framework alpha with Codex CLI
- URL: https://www.realagentcases.com/use-cases/upgrade-a-plugin-to-a-new-framework-alpha-with-codex-cli/
- Outcome: A one-line `codex exec` loop read the Datasette 1.0a20 upgrade docs, fixed the failing tests, and got the datasette-checkbox plugin passing again.
- Agents: OpenAI Codex CLI
- Category: Coding
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://simonwillison.net/2025/Nov/6/upgrading-datasette-plugins/ (2025-11-06)
- Last verified: 2026-10-15

### Build a terminal-to-HTML sharing tool with Claude Code for web
- URL: https://www.realagentcases.com/use-cases/build-a-terminal-to-html-sharing-tool-with-claude-code-for-web/
- Outcome: Claude Code for web built a terminal-to-html tool that turns pasted terminal output into a shareable gist-preview link, recorded start to finish in an 11-minute video.
- Agents: Claude Code for web
- Category: Coding
- Evidence: Shown on video
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://simonwillison.net/2025/Oct/23/claude-code-for-web-video/ (2025-10-23)
- Last verified: 2026-10-15

### Get DeepSeek-OCR running on an NVIDIA Spark with Claude Code
- URL: https://www.realagentcases.com/use-cases/get-deepseek-ocr-running-on-an-nvidia-spark-with-claude-code/
- Outcome: With four prompts and about 40 minutes, Claude Code figured out the right ARM CUDA PyTorch wheels, got DeepSeek-OCR running on NVIDIA Spark hardware, and OCR'd a test image.
- Agents: Claude Code
- Category: Coding
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://simonwillison.net/2025/Oct/20/deepseek-ocr-claude-code/ (2025-10-20)
- Last verified: 2026-10-15

### Write and ship a complete Datasette plugin with Claude Code
- URL: https://www.realagentcases.com/use-cases/write-and-ship-a-complete-datasette-plugin-with-claude-code/
- Outcome: Claude Code (Sonnet 4.5) wrote every line of code and every test for the datasette-os-info plugin with minimal prompting, and the author shipped it.
- Agents: Claude Code
- Category: Coding
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://simonwillison.net/2025/Oct/8/claude-datasette-plugins/ (2025-10-08)
- Last verified: 2026-10-15

### Speed up feature delivery by moving a team to Cursor
- URL: https://www.realagentcases.com/use-cases/speed-up-feature-delivery-by-moving-a-team-to-cursor/
- Outcome: Security startup Jit moved its engineers from JetBrains to Cursor and, according to the article, delivered features 1.8–3x faster.
- Agents: Cursor
- Category: Coding
- Evidence: Article or interview
- Claimed result: 1.8x–3x productivity gains (Reported by LeadDev; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://leaddev.com/technical-direction/how-jit-overcame-developer-resistance-shift-cursor (2025-05-15)
- Last verified: 2026-10-15

### Generate a colophon page for a GitHub Pages site with Claude Code
- URL: https://www.realagentcases.com/use-cases/generate-a-colophon-page-for-a-github-pages-site-with-claude-code/
- Outcome: Claude Code wrote a script that builds a mobile-friendly colophon page listing every tool and its commit history, then wired it into the GitHub Pages deploy workflow.
- Agents: Claude Code
- Category: Coding
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://simonwillison.net/2025/Mar/11/using-llms-for-code/ (2025-03-11)
- Last verified: 2026-10-15

### Vibe-code a 3D town-building game with Claude Code
- URL: https://www.realagentcases.com/use-cases/vibe-code-a-3d-town-building-game-with-claude-code/
- Outcome: From a single informal prompt, Claude Code built a working 3D game for placing buildings and driving around in about four minutes; follow-up prompts added a fire truck, traffic, and fires.
- Agents: Claude Code
- Category: Coding
- Evidence: Article or interview
- Claimed result: Working game in about four minutes (Reported by One Useful Thing; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.oneusefulthing.org/p/speaking-things-into-existence (2025-03-11)
- Last verified: 2026-10-15

## Content

### Write and publish a short podcast episode with Claude Code (needed edits)
- URL: https://www.realagentcases.com/use-cases/write-and-publish-a-short-podcast-episode-with-claude-code-needed-edits/
- Outcome: Claude Code wrote a short podcast script that the poster voiced with ElevenLabs and published via Anchor; the narration was clear but the script missed nuance and needed manual tweaks, so the poster rated it useful for drafts only.
- Agents: Claude Code
- Category: Content
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Partial. The script missed nuance and needed manual edits; the poster called it useful for drafts only.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/AIorDIY/status/2052464770406854825 (2026-05-07)
- Last verified: 2026-10-16

### Make years of calls, podcasts and posts searchable with Claude Code
- URL: https://www.realagentcases.com/use-cases/make-years-of-calls-podcasts-and-posts-searchable-with-claude-code/
- Outcome: Claude Code organized years of calls, podcasts, tweets and presentations into a searchable database connected to AI, with a daily job that adds new content; asked for a masterclass outline, it drew on 224 chunks from 78 sources.
- Agents: Claude Code
- Category: Content
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/andrehaykaljr/status/2021676546218434695 (2026-02-11)
- Last verified: 2026-10-16

### Turn one video script into a newsletter, titles, thread and description with Claude Code subagents
- URL: https://www.realagentcases.com/use-cases/turn-one-video-script-into-a-newsletter-titles-thread-and-description-with-claud/
- Outcome: From one long-form video script, an orchestrator in Claude Code calls subagents that write a newsletter, generate titles, create thumbnails, write a thread and draft the YouTube description.
- Agents: Claude Code
- Category: Content
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/moritzkremb/status/1951636932846645260 (2025-08-02)
- Last verified: 2026-10-16

### Batch-enter design prompts into an image generator: Operator vs Project Mariner
- URL: https://www.realagentcases.com/use-cases/batch-enter-design-prompts-into-an-image-generator-operator-vs-project-mariner/
- Outcome: For an Etsy design shop, Operator entered a ChatGPT-made list of prompts into Ideogram to generate draft designs after the user logged in; Project Mariner couldn't get past Ideogram's captcha even with the user in control.
- Agents: OpenAI Operator, Project Mariner
- Category: Content
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Partial. Operator worked once the user logged in; Project Mariner could not get past the captcha.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/CodyThieling/status/1927128568045183167 (2025-05-26)
- Last verified: 2026-10-16

### Turn podcast highlights into timestamped clips with Claude Cowork + Claude in Chrome + Codex
- URL: https://www.realagentcases.com/use-cases/turn-podcast-highlights-into-timestamped-clips-with-claude-cowork-claude-in-chro/
- Outcome: Claude Cowork expanded the poster's top 10 podcast takeaways, then used Claude in Chrome to find the transcript and timestamps for each clip, and Codex downloaded them.
- Agents: Claude Cowork, Claude in Chrome, OpenAI Codex
- Category: Content
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/daniel_mac8/status/2035782226571305039 (2026-03-22)
- Last verified: 2026-10-15

### Break writer's block on a book with Claude Code interviews
- URL: https://www.realagentcases.com/use-cases/break-writer-s-block-on-a-book-with-claude-code-interviews/
- Outcome: Claude Code reviewed a book draft, interviewed the author to fill gaps, and adapted his dictated answers into the manuscript in his own voice.
- Agents: Claude Code
- Category: Content
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/marckohlbrugge/status/2022992628107870469 (2026-02-15)
- Last verified: 2026-10-15

### Debrief an 84-minute YouTube panel with Claude Code
- URL: https://www.realagentcases.com/use-cases/debrief-an-84-minute-youtube-panel-with-claude-code/
- Outcome: Claude Code installed yt-dlp, pulled and cleaned the captions of an 84-minute Gurley/Gerstner/Nadella talk, wrote focused debriefs, then packaged it as a reusable skill.
- Agents: Claude Code
- Category: Content
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/vha14/status/2014848771503620499 (2026-01-23)
- Last verified: 2026-10-15

### Generate weekly AI content ideas with an n8n agent
- URL: https://www.realagentcases.com/use-cases/generate-weekly-ai-content-ideas-with-an-n8n-agent/
- Outcome: An n8n agent researches top AI news via Brave Search, generates content from templates, and emails a weekly idea pack.
- Agents: n8n agent
- Category: Content
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/_juliettech/status/1983231404914929686 (2025-10-28)
- Last verified: 2026-10-15

### Turn stroller-walk voice notes into a published article with Claude Code
- URL: https://www.realagentcases.com/use-cases/turn-stroller-walk-voice-notes-into-a-published-article-with-claude-code/
- Outcome: A non-developer fed rambling voice notes to Claude Code, which organized them into research themes, wrote a full article in her voice and made LinkedIn versions.
- Agents: Claude Code
- Category: Content
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://www.lennysnewsletter.com/p/everyone-should-be-using-claude-code (2025-10-14)
- Last verified: 2026-10-15

### Prep podcast research packets with an n8n agent
- URL: https://www.realagentcases.com/use-cases/prep-podcast-research-packets-with-an-n8n-agent/
- Outcome: Emily Lambert built an app on an n8n agent that researches a podcast topic, pulls news and tweets, then emails a script/outline to the team.
- Agents: n8n agent
- Category: Content
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/emilylambert/status/1932083209380274313 (2025-06-09)
- Last verified: 2026-10-15

### Turn a blog post into a Quarto slide deck with Claude Code
- URL: https://www.realagentcases.com/use-cases/turn-a-blog-post-into-a-quarto-slide-deck-with-claude-code/
- Outcome: Claude Code read one of the author's blog posts and turned it into a Quarto slide deck in three prompts.
- Agents: Claude Code
- Category: Content
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/kyle_e_walker/status/1931850344855691498 (2025-06-08)
- Last verified: 2026-10-15

## Data

### Turn a year of bank statements into a business expense spreadsheet with Claude Cowork
- URL: https://www.realagentcases.com/use-cases/turn-a-year-of-bank-statements-into-a-business-expense-spreadsheet-with-claude-c/
- Outcome: Claude Cowork extracted and categorized business expenses from a year of statements across several accounts into a well-organized spreadsheet; after some back-and-forth the poster fixed a handful of items by hand.
- Agents: Claude Cowork
- Category: Data
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/MikeMoves/status/2036708713562513630 (2026-03-25)
- Last verified: 2026-10-16

### Find smarter buying patterns in office supply order history with Claude Code
- URL: https://www.realagentcases.com/use-cases/find-smarter-buying-patterns-in-office-supply-order-history-with-claude-code/
- Outcome: Alex Hillman had Claude Code go through Indy Hall's paper products order history to suggest buying patterns that avoid costly last-minute orders.
- Agents: Claude Code
- Category: Data
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/alexhillman/status/2020352927848923387 (2026-02-08)
- Last verified: 2026-10-16

### Analyze seven years of credit card statements with Claude Cowork
- URL: https://www.realagentcases.com/use-cases/analyze-seven-years-of-credit-card-statements-with-claude-cowork/
- Outcome: From seven years of monthly credit card statement PDFs in zip files, Claude Cowork returned an Excel file with 5,726 extracted, categorized transactions plus analysis; issues were vendor transaction IDs and some miscategorized vendors it wasn't told how to handle.
- Agents: Claude Cowork
- Category: Data
- Evidence: Screenshots or logs
- Claimed result: 5,726 transactions categorized (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/_simonsmith/status/2018380086311800948 (2026-02-02)
- Last verified: 2026-10-16

### Let a whole team self-serve data via an agent-written schema guide
- URL: https://www.realagentcases.com/use-cases/let-a-whole-team-self-serve-data-via-an-agent-written-schema-guide/
- Outcome: Claude Code wrote a markdown guide to the company's data model and business terms from SQLMesh context and past queries; Hex's agent now answers the team's data requests, and the analyst hire was dropped.
- Agents: Claude Code, Hex agent
- Category: Data
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/dylangans/status/2047768695678578723 (2026-04-24)
- Last verified: 2026-10-15

### Merge provider price CSVs into one comparison table with Claude Code
- URL: https://www.realagentcases.com/use-cases/merge-provider-price-csvs-into-one-comparison-table-with-claude-code/
- Outcome: Claude Code read several local CSVs of model prices from different providers and produced one organized comparison table grouped by provider.
- Agents: Claude Code
- Category: Data
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/S0laceGao/status/2036850598855766030 (2026-03-25)
- Last verified: 2026-10-15

### Automate a daily analytics summary to Slack with Claude Code
- URL: https://www.realagentcases.com/use-cases/automate-a-daily-analytics-summary-to-slack-with-claude-code/
- Outcome: Claude Code set up a GitHub Actions cron that pulls PostHog data at 8am, summarizes it in Python and posts to Slack; the first report arrived the next morning.
- Agents: Claude Code
- Category: Data
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/PranavDate/status/2031210925806530779 (2026-03-10)
- Last verified: 2026-10-15

### Turn X bookmarks and likes into a searchable knowledge base
- URL: https://www.realagentcases.com/use-cases/turn-x-bookmarks-and-likes-into-a-searchable-knowledge-base/
- Outcome: The poster scraped all their X bookmarks and likes with Claude in Chrome and other tools, cleaned the linked pages with Firecrawl, and built a RAG chatbot over everything they had saved.
- Agents: Claude in Chrome, Vercel Agent Browser
- Category: Data
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/sheeki03/status/2013898161782481048 (2026-01-21)
- Last verified: 2026-10-15

### Analyze every entry of a huge JSONL file with Claude Code subagents
- URL: https://www.realagentcases.com/use-cases/analyze-every-entry-of-a-huge-jsonl-file-with-claude-code-subagents/
- Outcome: Claude Code previewed the data, trimmed fields, defined a schema and ran hundreds of Haiku subagents over ~45 minutes so every entry in a large JSONL file was analyzed.
- Agents: Claude Code
- Category: Data
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/sean_moriarity/status/2012220803023073539 (2026-01-16)
- Last verified: 2026-10-15

### Compare Claude Code, Codex and Gemini CLI on a 97k-message chat dataset
- URL: https://www.realagentcases.com/use-cases/compare-claude-code-codex-and-gemini-cli-on-a-97k-message-chat-dataset/
- Outcome: The poster gave the same 97k-message WeChat dataset to Claude Code, Codex and Gemini CLI as analysts and published results with an evaluation table.
- Agents: Claude Code, OpenAI Codex CLI, Gemini CLI
- Category: Data
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/tylerxiety/status/1988989909429022850 (2025-11-13)
- Last verified: 2026-10-15

### Reproduce an economics paper's findings from replication data with Claude
- URL: https://www.realagentcases.com/use-cases/reproduce-an-economics-paper-s-findings-from-replication-data-with-claude/
- Outcome: Given only the paper and its replication archive, Claude Sonnet 4.5 converted the STATA code to Python and reproduced the findings; the author spot-checked the results and GPT-5 Pro confirmed them.
- Agents: Claude (Sonnet 4.5)
- Category: Data
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.oneusefulthing.org/p/real-ai-agents-and-real-work (2025-09-29)
- Last verified: 2026-10-15

## Marketing

### Write a drama-style ad script modeled on a winning ad with Manus
- URL: https://www.realagentcases.com/use-cases/write-a-drama-style-ad-script-modeled-on-a-winning-ad-with-manus/
- Outcome: Manus wrote a drama-format script for one of the brand's top historical winning ad concepts, modeled on a winning drama ad from a beauty brand; the poster says it hit 5 ROAS right away.
- Agents: Manus
- Category: Marketing
- Evidence: Described by poster
- Claimed result: 5 ROAS off the bat (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/thankyouecom/status/2107259289224053025 (2026-10-05)
- Last verified: 2026-10-16

### Find car brands without a web app and email them a pitch with Manus
- URL: https://www.realagentcases.com/use-cases/find-car-brands-without-a-web-app-and-email-them-a-pitch-with-manus/
- Outcome: Manus searched for car brands without a web app, wrote an email and sent it to them; some replied and the poster built one an app.
- Agents: Manus
- Category: Marketing
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/JerryJenasom/status/2102306956203442534 (2026-09-22)
- Last verified: 2026-10-16

### Manage Google Ads with Claude Code connected to analytics and the website
- URL: https://www.realagentcases.com/use-cases/manage-google-ads-with-claude-code-connected-to-analytics-and-the-website/
- Outcome: With Google Ads, Google Analytics and the Webflow site connected to a Claude Code project, the poster adds negative keywords quickly, tracks ad performance, and found that mobile wasn't converting because the page showed nothing relevant above the fold.
- Agents: Claude Code
- Category: Marketing
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/michaelsxacc/status/2087519918803566662 (2026-08-12)
- Last verified: 2026-10-16

### Qualify scraped leads against your ideal customer before cold email with Claude Code
- URL: https://www.realagentcases.com/use-cases/qualify-scraped-leads-against-your-ideal-customer-before-cold-email-with-claude/
- Outcome: Claude Code reads each scraped company's website and checks it against the poster's ideal customer profile, dropping about 30% of leads before outreach; the poster says reply rates roughly doubled and they booked 100+ meetings in 3 months.
- Agents: Claude Code
- Category: Marketing
- Evidence: Described by poster
- Claimed result: 100+ meetings booked in 3 months (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/ZaidAD_/status/2062271127213236350 (2026-06-03)
- Last verified: 2026-10-16

### Submit a product to ~300 directories and launch sites with Claude in Chrome
- URL: https://www.realagentcases.com/use-cases/submit-a-product-to-300-directories-and-launch-sites-with-claude-in-chrome/
- Outcome: In one day, Claude in Chrome filled directory and launch-platform forms 5 tabs at a time; 134 of about 300 submissions went through, with the human handling logins and captchas.
- Agents: Claude in Chrome
- Category: Marketing
- Evidence: Described by poster
- Claimed result: 134 submissions went through (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/_ksl_dev/status/2105293870766444553 (2026-09-30)
- Last verified: 2026-10-15

### Fix a Product Hunt X handle across launches and comments with a browser agent
- URL: https://www.realagentcases.com/use-cases/fix-a-product-hunt-x-handle-across-launches-and-comments-with-a-browser-agent/
- Outcome: A browser agent corrected the poster's X handle from johnrushx to johnrush across all Product Hunt launches and hundreds of comments.
- Agents: browser agent (Claude in Chrome)
- Category: Marketing
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/johnrush/status/2087198876688412929 (2026-08-11)
- Last verified: 2026-10-15

### Audit 90 days of Meta ads from a CSV export with Claude Code
- URL: https://www.realagentcases.com/use-cases/audit-90-days-of-meta-ads-from-a-csv-export-with-claude-code/
- Outcome: Claude Code flagged 14 money-losing creatives, found that 7 of the top 10 ads opened with a question, and spotted 3 ad sets cannibalizing each other.
- Agents: Claude Code
- Category: Marketing
- Evidence: Described by poster
- Claimed result: 14 money-losing creatives flagged (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://x.com/propialis/status/2041264181106692318 (2026-04-06)
- Last verified: 2026-10-15

### Automate a daily cold-email pipeline with Claude Code
- URL: https://www.realagentcases.com/use-cases/automate-a-daily-cold-email-pipeline-with-claude-code/
- Outcome: A weekend-built Claude Code pipeline scrapes leads, scores them against the ICP, writes spintax copy and launches the Instantly campaign with one command; the poster says it saves about 2 hours each morning.
- Agents: Claude Code
- Category: Marketing
- Evidence: Described by poster
- Claimed result: 2 hours back every morning (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/Jason_scales1/status/2039976196339859727 (2026-04-03)
- Last verified: 2026-10-15

### Build a B2B SaaS press list of 40–50 journalists with Manus
- URL: https://www.realagentcases.com/use-cases/build-a-b2b-saas-press-list-of-40-50-journalists-with-manus/
- Outcome: Manus produced a Notion-ready press list of journalists covering mid-market SaaS/enterprise software with contacts, recent articles and pitch angles; the author says work that used to take ~6 hours took about 25 minutes, with some contact verification still needed.
- Agents: Manus
- Category: Marketing
- Evidence: Article or interview
- Claimed result: 6 hours of work done in 25 minutes (Reported by The AI Enterprise; a claim from the source, not independently verified)
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.theaienterprise.io/p/ai-agents-research (2025-10-21)
- Last verified: 2026-10-15

### Generate 60+ AI-consulting sales prospects with Manus
- URL: https://www.realagentcases.com/use-cases/generate-60-ai-consulting-sales-prospects-with-manus/
- Outcome: A Manus prompt for mid-market AI-implementation prospects returned about 73 qualified companies with decision-maker contacts and outreach notes in a Notion/spreadsheet export.
- Agents: Manus
- Category: Marketing
- Evidence: Article or interview
- Claimed result: 73 prospects found (Reported by The AI Enterprise; a claim from the source, not independently verified)
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.theaienterprise.io/p/ai-agents-research (2025-10-21)
- Last verified: 2026-10-15

### Find Southeast AI events for meetup promotion with Manus
- URL: https://www.realagentcases.com/use-cases/find-southeast-ai-events-for-meetup-promotion-with-manus/
- Outcome: Manus found 34 upcoming AI/tech events in North Carolina and the Southeast suitable for promoting a meetup, exported to Notion with contacts and opportunity type, cutting an afternoon of work to about 20 minutes.
- Agents: Manus
- Category: Marketing
- Evidence: Article or interview
- Claimed result: 34 relevant events found (Reported by The AI Enterprise; a claim from the source, not independently verified)
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.theaienterprise.io/p/ai-agents-research (2025-10-21)
- Last verified: 2026-10-15

### Weekly X analytics review and growth coaching with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/weekly-x-analytics-review-and-growth-coaching-with-chatgpt-agent/
- Outcome: Every Monday ChatGPT agent pulls the poster's X analytics, studies which posts worked, compares profile visits with follower growth, and suggests bio and pinned-post changes.
- Agents: ChatGPT agent
- Category: Marketing
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/AIwithJai/status/1967708746752327852 (2025-09-15)
- Last verified: 2026-10-15

### Fix Google Ads conversion tracking with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/fix-google-ads-conversion-tracking-with-chatgpt-agent/
- Outcome: A solo creator had ChatGPT agent log into her Google Ads account and fix a conversion-tracking issue.
- Agents: ChatGPT agent
- Category: Marketing
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/evielync/status/1953929013703193013 (2025-08-08)
- Last verified: 2026-10-15

### Build an SEO content calendar with Manus
- URL: https://www.realagentcases.com/use-cases/build-an-seo-content-calendar-with-manus/
- Outcome: Manus studied the author's products and writing, researched SEO and keywords, and produced a content calendar in a 30+ minute run that the author says cost about $10.
- Agents: Manus
- Category: Marketing
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.asad.pw/ai-agents-experiences-with-manus-operator-and-more/ (2025-05-31)
- Last verified: 2026-10-15

### Generate qualified sales leads with Zapier Agents
- URL: https://www.realagentcases.com/use-cases/generate-qualified-sales-leads-with-zapier-agents/
- Outcome: According to Zapier's customer story, Slate Studios produced 2,000+ qualified leads in one month with a single Zapier Agent that also handled data reformatting and competitive research.
- Agents: Zapier Agents
- Category: Marketing
- Evidence: Company case study
- Claimed result: 2,000+ leads generated (According to Zapier; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://zapier.com/blog/how-slate-generates-2000-leads-with-zapier-agents/ (2025-04-30)
- Last verified: 2026-10-15

### Automate PR lead discovery from news with Zapier Agents
- URL: https://www.realagentcases.com/use-cases/automate-pr-lead-discovery-from-news-with-zapier-agents/
- Outcome: According to Zapier's customer story, NisonCo grew weekly leads from 270 to nearly 400 (+48%) and saved about $30,000 a year with an agent that scans news for new businesses in its target industries.
- Agents: Zapier Agents
- Category: Marketing
- Evidence: Company case study
- Claimed result: 48% increase in leads (According to Zapier; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://zapier.com/blog/how-nisonco-fuels-business-growth-with-zapier-agents/ (2025-02-18)
- Last verified: 2026-10-15

## Ops

### Build a Power Automate Desktop flow with Claude Cowork
- URL: https://www.realagentcases.com/use-cases/build-a-power-automate-desktop-flow-with-claude-cowork/
- Outcome: After Copilot suggested a subflow that didn't exist in Power Automate Desktop, the poster asked Claude Cowork, which took over and built the flow on their machine.
- Agents: Claude Cowork
- Category: Ops
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/magicj3/status/2105499283126407656 (2026-10-01)
- Last verified: 2026-10-16

### Submit a hosting provider bug report from log warnings with Claude in Chrome
- URL: https://www.realagentcases.com/use-cases/submit-a-hosting-provider-bug-report-from-log-warnings-with-claude-in-chrome/
- Outcome: Using the poster's logged-in browser, Claude in Chrome submitted a bug report to DigitalOcean after finding warnings in the logs.
- Agents: Claude in Chrome
- Category: Ops
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/ndbroadbent/status/2104499860368138397 (2026-09-28)
- Last verified: 2026-10-16

### Split, rename and sort about 200 scanned invoices with Claude Cowork
- URL: https://www.realagentcases.com/use-cases/split-rename-and-sort-about-200-scanned-invoices-with-claude-cowork/
- Outcome: After the poster scanned almost 200 invoices, Claude Cowork split them out, renamed them, sorted them into folders and created a summary Excel file.
- Agents: Claude Cowork
- Category: Ops
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/IlyaPion/status/2102783493482709261 (2026-09-23)
- Last verified: 2026-10-16

### Find, negotiate and review a small office lease with Codex
- URL: https://www.realagentcases.com/use-cases/find-negotiate-and-review-a-small-office-lease-with-codex/
- Outcome: Codex found a 74 sq ft commercial lease near the poster's kids' school, negotiated the terms and reviewed the contract; the poster found it cheaper than a coworking desk.
- Agents: Codex
- Category: Ops
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/kolencherry/status/2094983316306858030 (2026-09-02)
- Last verified: 2026-10-16

### Prepare an Indian income tax return (ITR-3) with Claude Code
- URL: https://www.realagentcases.com/use-cases/prepare-an-indian-income-tax-return-itr-3-with-claude-code/
- Outcome: Given tax documents and a year of bank statements, Claude Code reconciled the figures against the poster's AIS (annual information statement), caught 2 mistakes that could have triggered notices, and computed the tax; the poster then filed.
- Agents: Claude Code
- Category: Ops
- Evidence: Described by poster
- Claimed result: 2 mistakes caught (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/m_jamal_jm/status/2075517698319286312 (2026-07-10)
- Last verified: 2026-10-16

### Reconcile crypto trades across chains and post them to Xero with Claude Code
- URL: https://www.realagentcases.com/use-cases/reconcile-crypto-trades-across-chains-and-post-them-to-xero-with-claude-code/
- Outcome: Claude Code tracked on-chain swaps across 10+ assets, produced a monthly gain/loss breakdown with reconciled exchange rates, then built a Xero connection and posted journal entries.
- Agents: Claude Code
- Category: Ops
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/ksaitor/status/2042163099600638171 (2026-04-09)
- Last verified: 2026-10-16

### Match a tax-prep checklist against a receipts folder with Claude Cowork
- URL: https://www.realagentcases.com/use-cases/match-a-tax-prep-checklist-against-a-receipts-folder-with-claude-cowork/
- Outcome: The poster says matching a spreadsheet of items for their accountant against a folder of receipts usually took 4–6 hours, and that with Claude Cowork it took about an hour.
- Agents: Claude Cowork
- Category: Ops
- Evidence: Described by poster
- Claimed result: 4–6 hour job done in about an hour (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/bharat/status/2034068158123020569 (2026-03-18)
- Last verified: 2026-10-16

### Prepare and file overdue German VAT returns with Claude Cowork and Claude in Chrome
- URL: https://www.realagentcases.com/use-cases/prepare-and-file-overdue-german-vat-returns-with-claude-cowork-and-claude-in-chr/
- Outcome: From a folder of invoice and receipt PDFs, Claude Cowork prepared quarterly accounts; after the poster logged in to the Elster tax portal, Claude in Chrome found, filled and submitted the Q3 2025 VAT forms.
- Agents: Claude Cowork, Claude in Chrome
- Category: Ops
- Evidence: Shown on video
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/EastlondonDev/status/2020826097890242582 (2026-02-09)
- Last verified: 2026-10-16

### Turn six months of receipts into a categorized spreadsheet with Claude Cowork
- URL: https://www.realagentcases.com/use-cases/turn-six-months-of-receipts-into-a-categorized-spreadsheet-with-claude-cowork/
- Outcome: Pointed at a folder, Claude Cowork organized 6 months of receipts into a categorized spreadsheet with monthly breakdowns.
- Agents: Claude Cowork
- Category: Ops
- Evidence: Shown on video
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/ChuckYouSuck/status/2014502869336989810 (2026-01-23)
- Last verified: 2026-10-16

### Do year-end bookkeeping and Xero reconciliation with Claude Code
- URL: https://www.realagentcases.com/use-cases/do-year-end-bookkeeping-and-xero-reconciliation-with-claude-code/
- Outcome: Claude Code categorized a folder of receipts, cross-referenced exported bank CSVs against Xero to find duplicates and missing transactions, used the browser extension to report on reconciliation items, and produced a cost-cutting report; the poster says this replaced roughly $2–3K of accounting-firm work.
- Agents: Claude Code, Claude in Chrome
- Category: Ops
- Evidence: Screenshots or logs
- Claimed result: ~2–3K accounting-firm fee avoided (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/amirmxt/status/2004658010212499713 (2025-12-26)
- Last verified: 2026-10-16

### Complete a 31-step web form: ChatGPT agent vs Perplexity Comet
- URL: https://www.realagentcases.com/use-cases/complete-a-31-step-web-form-chatgpt-agent-vs-perplexity-comet/
- Outcome: On a 31-step form with buttons, dropdowns and fields, Comet failed after 3–5 steps, while ChatGPT agent completed all 31 and corrected its own mistakes by checking against the steps.
- Agents: ChatGPT agent, Perplexity Comet
- Category: Ops
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Partial. Comet failed after 3–5 steps; ChatGPT agent finished all 31.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/rohitdotmittal/status/1951450652867174754 (2025-08-02)
- Last verified: 2026-10-16

### Enter retail-arbitrage receipts into a spreadsheet with Claude Cowork
- URL: https://www.realagentcases.com/use-cases/enter-retail-arbitrage-receipts-into-a-spreadsheet-with-claude-cowork/
- Outcome: Claude Cowork weekly processes a folder of RA receipts into a spreadsheet with buy cost, store and item; the poster still double-checks but says admin time is way down.
- Agents: Claude Cowork
- Category: Ops
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/ZachAltmyer/status/2033904740359147632 (2026-03-17)
- Last verified: 2026-10-15

### Draft government-solicitation proposals in 25 minutes with n8n AI workflows
- URL: https://www.realagentcases.com/use-cases/draft-government-solicitation-proposals-in-25-minutes-with-n8n-ai-workflows/
- Outcome: According to n8n's case study, Field Aerospace's self-hosted n8n AI workflows bring a proposal draft that used to take about two weeks of work from three or four people to an 80% starting point in roughly 25 minutes, and replaced two legacy tools, eliminating approximately $30,000 in annual software costs (the case study's headline gives $22,000 a year).
- Agents: n8n (AI-powered workflows)
- Category: Ops
- Evidence: Company case study
- Claimed result: Proposal time cut from 2 weeks to 25 minutes (According to n8n; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://n8n.io/case-studies/field-aerospace/ (2026-03-02)
- Last verified: 2026-10-15

### Build a weekly business KPI dashboard workflow with Claude Code + n8n
- URL: https://www.realagentcases.com/use-cases/build-a-weekly-business-kpi-dashboard-workflow-with-claude-code-n8n/
- Outcome: Claude Code built a complex n8n workflow for a client's weekly KPI dashboard in about 15 minutes; the workflow then runs unattended on n8n.
- Agents: Claude Code, n8n
- Category: Ops
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://landauai.com/ai-agents-claude-code-vs-workflow-tools-n8n-which-should-automate-your-processes/ (2026-02-27)
- Last verified: 2026-10-15

### Auto-coach sales reps from call recordings with Zapier Agents
- URL: https://www.realagentcases.com/use-cases/auto-coach-sales-reps-from-call-recordings-with-zapier-agents/
- Outcome: According to Zapier's customer story, Healthie's call-summary and coaching agents save each of about 20 reps and CSMs 2–3 hours a week, more than 60 hours a week across the team.
- Agents: Zapier Agents
- Category: Ops
- Evidence: Company case study
- Claimed result: 60+ hours a week saved (According to Zapier; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://zapier.com/blog/healthie-saves-60-hours-per-week-with-ai-agents/ (2025-11-25)
- Last verified: 2026-10-15

### Sort and rename messy invoices for taxes with Claude Code
- URL: https://www.realagentcases.com/use-cases/sort-and-rename-messy-invoices-for-taxes-with-claude-code/
- Outcome: Claude Code reads each file in a messy invoice folder, renames it to 'YYYY-MM-DD Vendor - Invoice - ProductOrService.pdf' and files it in the right folder.
- Agents: Claude Code
- Category: Ops
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://www.lennysnewsletter.com/p/everyone-should-be-using-claude-code (2025-10-14)
- Last verified: 2026-10-15

### Categorize 419 ledger transactions into an Excel template with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/categorize-419-ledger-transactions-into-an-excel-template-with-chatgpt-agent/
- Outcome: Given an Excel ledger template and 419 transactions, ChatGPT agent asked for help on ambiguous merchants, then keyed in all ~400 rows and took 4 minutes.
- Agents: ChatGPT agent
- Category: Ops
- Evidence: Screenshots or logs
- Claimed result: Done in 4 minutes (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/jasononfirms/status/1963993526515192040 (2025-09-05)
- Last verified: 2026-10-15

### Allocate a 30-employee payroll journal across departments with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/allocate-a-30-employee-payroll-journal-across-departments-with-chatgpt-agent/
- Outcome: ChatGPT agent mapped a payroll CSV into a 30-employee allocation spreadsheet across 6 departments, making 180 entries and summing employer-paid columns in under 4 minutes.
- Agents: ChatGPT agent
- Category: Ops
- Evidence: Screenshots or logs
- Claimed result: 180 entries in under 4 minutes (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/jasononfirms/status/1963628292873109898 (2025-09-04)
- Last verified: 2026-10-15

### Fill an accelerator application with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/fill-an-accelerator-application-with-chatgpt-agent/
- Outcome: ChatGPT agent filled an accelerator application using the poster's context and chat memory; the poster rated it 8/10, and also had it research vendors into a spreadsheet (7/10).
- Agents: ChatGPT agent
- Category: Ops
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/mahtabsoin/status/1949203910922764689 (2025-07-26)
- Last verified: 2026-10-15

### Diagnose a Kubernetes pod-scheduling outage with Claude Code
- URL: https://www.realagentcases.com/use-cases/diagnose-a-kubernetes-pod-scheduling-outage-with-claude-code/
- Outcome: Anthropic's Data Infrastructure team gave Claude Code dashboard screenshots, and it traced the problem to pod IP exhaustion and supplied the commands to add a new IP pool; Anthropic says this saved about 20 minutes during an outage.
- Agents: Claude Code
- Category: Ops
- Evidence: Company case study
- Claimed result: 20 minutes saved during an outage (According to Anthropic; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://www.anthropic.com/news/how-anthropic-teams-use-claude-code (2025-07-24)
- Last verified: 2026-10-15

### Three-way match purchase orders against delivery and invoice docs with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/three-way-match-purchase-orders-against-delivery-and-invoice-docs-with-chatgpt-a/
- Outcome: In about 4 minutes ChatGPT agent checked that all 5 POs had matching delivery and invoice documents with no variances, combined the PDFs, and wrote a short report.
- Agents: ChatGPT agent
- Category: Ops
- Evidence: Screenshots or logs
- Claimed result: Done in around 4 minutes (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/jand__/status/1946283011445784991 (2025-07-18)
- Last verified: 2026-10-15

## Personal

### Log new board game purchases and add them to BoardGameGeek with Claude in Chrome
- URL: https://www.realagentcases.com/use-cases/log-new-board-game-purchases-and-add-them-to-boardgamegeek-with-claude-in-chrome/
- Outcome: From one instruction, the agent found the ordered games, created notes, downloaded cover images, added metadata and added the games to the poster's BoardGameGeek account using Claude in Chrome.
- Agents: Claude in Chrome
- Category: Personal
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://x.com/dSebastien/status/2110255458934301089 (2026-10-14)
- Last verified: 2026-10-16

### Write the next gym training cycle from workout logs with Claude Code
- URL: https://www.realagentcases.com/use-cases/write-the-next-gym-training-cycle-from-workout-logs-with-claude-code/
- Outcome: Claude Code read four weeks of the poster's workout logs, found where lifts stalled, wrote the next training cycle with a reason for each starting weight, loaded the routines into the tracking app, and noted where the app would suggest the wrong weight.
- Agents: Claude Code
- Category: Personal
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/Moretti_ale90/status/2106692994942308605 (2026-10-04)
- Last verified: 2026-10-16

### Rename a pile of train tickets by date and destination with Claude Cowork
- URL: https://www.realagentcases.com/use-cases/rename-a-pile-of-train-tickets-by-date-and-destination-with-claude-cowork/
- Outcome: Claude Cowork read a pile of train tickets and renamed each file by its correct date and destination.
- Agents: Claude Cowork
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/airocket2026/status/2104541855815991507 (2026-09-28)
- Last verified: 2026-10-16

### Turn an inbox into a daily to-do list with Claude Cowork (abandoned)
- URL: https://www.realagentcases.com/use-cases/turn-an-inbox-into-a-daily-to-do-list-with-claude-cowork-abandoned/
- Outcome: Claude Cowork summarized the poster's email into a to-do list, but the list grew long and unmanageable, so the poster stopped reading it and turned it off as a waste of tokens.
- Agents: Claude Cowork
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Failure. The to-do list grew too long to use, so the poster stopped reading it and turned it off.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/uliang6482/status/2101965199452610582 (2026-09-21)
- Last verified: 2026-10-16

### Shop for cheaper home insurance quotes with Claude Code and Chrome
- URL: https://www.realagentcases.com/use-cases/shop-for-cheaper-home-insurance-quotes-with-claude-code-and-chrome/
- Outcome: Using the poster's coverage details and access to Chrome, Gmail and the existing policy, Claude Code sought quotes from about a dozen insurers and found two strong candidates; the poster says they were approved by AAA at $3,100/year, a $4,600/year saving.
- Agents: Claude Code, Claude in Chrome
- Category: Personal
- Evidence: Described by poster
- Claimed result: $4,600/year saved (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/sethprattsf/status/2100023090860429659 (2026-09-16)
- Last verified: 2026-10-16

### Screen and clear an email inbox in one pass with Claude Code
- URL: https://www.realagentcases.com/use-cases/screen-and-clear-an-email-inbox-in-one-pass-with-claude-code/
- Outcome: The poster runs all email in Claude Code: in one pass it screened 27 emails, archived and unsubscribed, and sent sub-agents after the three that needed real thought.
- Agents: Claude Code
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/francoisdefitte/status/2086831437047566793 (2026-08-10)
- Last verified: 2026-10-16

### Organize a Spanish visa renewal and fill the forms with Claude and Claude in Chrome
- URL: https://www.realagentcases.com/use-cases/organize-a-spanish-visa-renewal-and-fill-the-forms-with-claude-and-claude-in-chr/
- Outcome: For a Spanish visa renewal, the poster set up a bureaucracy folder with Claude, fed in the family's information, had Claude verify documents, then used Claude in Chrome to fill the online forms faster and more accurately than checking by hand.
- Agents: Claude, Claude in Chrome
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/milanm_/status/2071962695264288973 (2026-06-30)
- Last verified: 2026-10-16

### Transcribe and organize 411 family voice memos with Claude Code
- URL: https://www.realagentcases.com/use-cases/transcribe-and-organize-411-family-voice-memos-with-claude-code/
- Outcome: Claude Code transcribed and clustered 411 voice memos left by the poster's late grandmother and built a small site organized by mood and date, which the family uses more than the photo album.
- Agents: Claude Code
- Category: Personal
- Evidence: Screenshots or logs
- Claimed result: 411 voice memos transcribed (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/the_smart_ape/status/2051926703287533820 (2026-05-06)
- Last verified: 2026-10-16

### Book a doctor's appointment with agent mode (blocked)
- URL: https://www.realagentcases.com/use-cases/book-a-doctor-s-appointment-with-agent-mode-blocked/
- Outcome: The poster tried to set up a doctor's appointment with agent mode and got nowhere; they say ChatGPT and Manus agents are held back by blocking and captchas.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Failure. Got nowhere; the poster says blocking and captchas stopped it.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/NanjuanS/status/2025030709887512754 (2026-02-21)
- Last verified: 2026-10-16

### Draft a firm landlord notice about a repeat water leak with Claude Code
- URL: https://www.realagentcases.com/use-cases/draft-a-firm-landlord-notice-about-a-repeat-water-leak-with-claude-code/
- Outcome: Instead of paying a lawyer for a strongly worded notice, the poster gave Claude Code the lease documents and context about a second water leak; it drafted an email that got them the concessions they wanted.
- Agents: Claude Code
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/4484/status/2024645227089428607 (2026-02-20)
- Last verified: 2026-10-16

### Apply to LinkedIn jobs from a resume with Perplexity Comet
- URL: https://www.realagentcases.com/use-cases/apply-to-linkedin-jobs-from-a-resume-with-perplexity-comet/
- Outcome: Comparing Comet and ChatGPT Atlas for LinkedIn job applications, the poster found Comet far better; with a detailed prompt and their resume, it applied to 20+ jobs and sent notes to HR.
- Agents: Perplexity Comet
- Category: Personal
- Evidence: Screenshots or logs
- Claimed result: 20+ job applications (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/vedanthpeace/status/1982463132980392390 (2025-10-26)
- Last verified: 2026-10-16

### Spot when you avoid conflict in meeting recordings with Claude Code
- URL: https://www.realagentcases.com/use-cases/spot-when-you-avoid-conflict-in-meeting-recordings-with-claude-code/
- Outcome: Dan Shipper puts all his meeting recordings in a folder and asks Claude Code to find every time he subtly avoided conflict.
- Agents: Claude Code
- Category: Personal
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://www.lennysnewsletter.com/p/everyone-should-be-using-claude-code (2025-10-14)
- Last verified: 2026-10-16

### Diagnose why your Mac is running slow with Claude Code
- URL: https://www.realagentcases.com/use-cases/diagnose-why-your-mac-is-running-slow-with-claude-code/
- Outcome: Anthony Roux uses Claude Code to check load, memory pressure, disk space, stuck processes and swap, dig into causes like caches, Docker and Time Machine snapshots, explain the results and suggest fixes with risk notes.
- Agents: Claude Code
- Category: Personal
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://www.lennysnewsletter.com/p/everyone-should-be-using-claude-code (2025-10-14)
- Last verified: 2026-10-16

### Notify Germany's TV licence (GEZ) office of a house move with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/notify-germany-s-tv-licence-office-of-a-house-move-with-chatgpt-agent/
- Outcome: After a few problems, ChatGPT agent found the GEZ (public broadcasting fee) contact form, wrote a message and entered the new address; the poster only had to solve one captcha.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/kimmonismus/status/1961927860538515829 (2025-08-30)
- Last verified: 2026-10-16

### Find a restaurant-week steak deal and book a table for 2 with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/find-a-restaurant-week-steak-deal-and-book-a-table-for-2-with-chatgpt-agent/
- Outcome: Asked to find the best Dine LA steak dinner deal and book for 2 the next day, ChatGPT agent booked Dear John's for 8:15, though it misread the menu by treating sides as a separate course.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://x.com/wittinglyai/status/1953481169707610268 (2025-08-07)
- Last verified: 2026-10-16

### Pay a UK car tax reminder with ChatGPT agent (handoff for payment)
- URL: https://www.realagentcases.com/use-cases/pay-a-uk-car-tax-reminder-with-chatgpt-agent-handoff-for-payment/
- Outcome: From a photo of the posted reminder and a chosen payment option, ChatGPT agent found the government website and filled out the forms; the poster took over to enter card details.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/pauldm/status/1950157691084517393 (2025-07-29)
- Last verified: 2026-10-16

### Apply to a job with ChatGPT agent (small-button trouble)
- URL: https://www.realagentcases.com/use-cases/apply-to-a-job-with-chatgpt-agent-small-button-trouble/
- Outcome: Testing ChatGPT agent on a real job application (to OpenAI), the poster found it much better than the earlier Operator but it struggled to click small buttons; they rated Manus as slightly ahead at times.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Shown on video
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Partial. Struggled to click small buttons during the application.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/affaan/status/1949007302193029325 (2025-07-26)
- Last verified: 2026-10-16

### Track and cancel unused subscriptions with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/track-and-cancel-unused-subscriptions-with-chatgpt-agent/
- Outcome: ChatGPT agent scans for recurring charges and reminds the author to cancel before renewal, helping cancel unused streaming services.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.tomsguide.com/ai/5-clever-ways-i-use-chatgpt-agent-to-save-serious-money-each-month (2025-10-10)
- Last verified: 2026-10-15

### Find the best flight options on Skyscanner with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/find-the-best-flight-options-on-skyscanner-with-chatgpt-agent/
- Outcome: ChatGPT agent searched Skyscanner across flexible dates and returned the flight the author would actually book in 3 of 3 tests, though airline sites and 3D Secure blocked it from completing the booking.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://wow.pjh.is/journal/can-chatgpt-agent-find-a-flight (2025-08-22)
- Last verified: 2026-10-15

### Find an available domain name with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/find-an-available-domain-name-with-chatgpt-agent/
- Outcome: The poster found ChatGPT agent useful for finding an available domain name that met a set of constraints.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/davlgd/status/1955070135833620532 (2025-08-12)
- Last verified: 2026-10-15

### Book a business-lunch restaurant reservation with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/book-a-business-lunch-restaurant-reservation-with-chatgpt-agent/
- Outcome: From a terse prompt, ChatGPT agent found a restaurant, filled in the form (including a shellfish allergy), and got a confirmed reservation in 11 minutes versus about 2 manually.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.nngroup.com/articles/impressions-chatgpt-agent/ (2025-08-08)
- Last verified: 2026-10-15

### Sign up for a Steam account with Comet Assistant
- URL: https://www.realagentcases.com/use-cases/sign-up-for-a-steam-account-with-comet-assistant/
- Outcome: Comet Assistant created a Steam account with the email the poster gave it, and solved a captcha the poster had failed.
- Agents: Perplexity Comet Assistant
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/sanib4r/status/1952765629343166863 (2025-08-05)
- Last verified: 2026-10-15

### Book a restaurant reservation with ChatGPT agent (failure case)
- URL: https://www.realagentcases.com/use-cases/book-a-restaurant-reservation-with-chatgpt-agent-failure-case/
- Outcome: ChatGPT agent took about 5 minutes just to reach the reservation page, then handed back; the poster called it a waste of time.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Failure. Took about 5 minutes just to reach the booking page, then handed it back; the poster called it a waste of time.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/AnthonyShoosh/status/1951524727605837921 (2025-08-02)
- Last verified: 2026-10-15

### Weekly meal plan, priced grocery list and spreadsheet with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/weekly-meal-plan-priced-grocery-list-and-spreadsheet-with-chatgpt-agent/
- Outcome: ChatGPT agent created five healthy dinner recipes and a grocery list sorted by store section, checked local prices via Instacart data, and exported everything to Google Sheets.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.tomsguide.com/ai/i-tested-chatgpt-agent-on-5-everyday-tasks-heres-what-happened (2025-07-28)
- Last verified: 2026-10-15

### Plan a house-inspection Saturday with ChatGPT agent (blocked)
- URL: https://www.realagentcases.com/use-cases/plan-a-house-inspection-saturday-with-chatgpt-agent-blocked/
- Outcome: ChatGPT agent was asked to search a real-estate site for 4-bedroom Melbourne homes, save listings and route Saturday inspections, but the site and agent blocked each other.
- Agents: ChatGPT agent
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Failure. The real-estate site and the agent blocked each other, so the search never happened.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: original
- Source: https://x.com/Hessamsh/status/1949742495539302703 (2025-07-28)
- Last verified: 2026-10-15

### Rename, organize and de-duplicate files with Claude Code
- URL: https://www.realagentcases.com/use-cases/rename-organize-and-de-duplicate-files-with-claude-code/
- Outcome: Claude Code organized and renamed the poster's files and folders and found and removed duplicates; the poster says it worked incredibly well.
- Agents: Claude Code
- Category: Personal
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/JoniYashvili/status/1948767932202238289 (2025-07-25)
- Last verified: 2026-10-15

### Organize 225 Apple Notes with Claude Code
- URL: https://www.realagentcases.com/use-cases/organize-225-apple-notes-with-claude-code/
- Outcome: Claude Code read and changed the local Apple Notes database and organized 225 notes in about 5 minutes with little manual effort.
- Agents: Claude Code
- Category: Personal
- Evidence: Described by poster
- Claimed result: 225 notes organized in about 5 minutes (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/kirjd/status/1943374929585672439 (2025-07-10)
- Last verified: 2026-10-15

### Build a parenting shift-tracker app with Replit Agent
- URL: https://www.realagentcases.com/use-cases/build-a-parenting-shift-tracker-app-with-replit-agent/
- Outcome: A non-coder built ParentShift, a working app for couples to track night-duty shifts, using Replit Agent; the author reports it took about 3 hours and cost under $50.
- Agents: Replit Agent
- Category: Personal
- Evidence: Screenshots or logs
- Claimed result: Idea to app in 3 hours for $50 (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.producingparadise.com/articles/workflows/from-idea-to-app-in-3-hours-for-50-how-i-used-ai-to-build-parentshift (2025-06-25)
- Last verified: 2026-10-15

## Research

### Extend a family tree two generations with Claude Code and browser agents
- URL: https://www.realagentcases.com/use-cases/extend-a-family-tree-two-generations-with-claude-code-and-browser-agents/
- Outcome: Working from the family's archive scans, Claude Code with memory and the Claude browser plugin cross-checked church books, ship manifests and naturalization records, built a 300-page family wiki, and found two more generations in a day; the family is now checking it for errors.
- Agents: Claude Code, Claude in Chrome, Perplexity Comet
- Category: Research
- Evidence: Shown on video
- Claimed result: 300-page family wiki (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/staskulesh/status/2036173711053651984 (2026-03-23)
- Last verified: 2026-10-16

### Find hard-to-locate company tax IDs and addresses with GPT-5 Pro
- URL: https://www.realagentcases.com/use-cases/find-hard-to-locate-company-tax-ids-and-addresses-with-gpt-5-pro/
- Outcome: Working for over 45 minutes, GPT-5 Pro found correct FEINs for 4 of 6 subsidiaries on the Florida Division of Corporations site and one address via a FINRA BrokerCheck report; the poster double-checked the FEINs.
- Agents: GPT-5 Pro
- Category: Research
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://x.com/deredleritt3r/status/1986099337303126390 (2025-11-05)
- Last verified: 2026-10-16

### Find links to content with Project Mariner (no links returned)
- URL: https://www.realagentcases.com/use-cases/find-links-to-content-with-project-mariner-no-links-returned/
- Outcome: Asked to find links to content, Project Mariner found the content but wouldn't return the links, which the poster said made the result not very helpful.
- Agents: Project Mariner
- Category: Research
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Failure. Found the content but would not return the links, which the poster said made it unhelpful.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/Indy_triguy/status/1925727943436669316 (2025-05-23)
- Last verified: 2026-10-16

### Mine Gmail for on-record PR quotes with Claude Cowork
- URL: https://www.realagentcases.com/use-cases/mine-gmail-for-on-record-pr-quotes-with-claude-cowork/
- Outcome: Connected read-only to Gmail, Claude Cowork found 12 relevant PR pitches in about a minute, then narrowed to eight on-record quotes with publish permission that the journalist verified before use.
- Agents: Claude Cowork
- Category: Research
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://tech.yahoo.com/ai/claude/articles/tried-claude-cowork-gmail-inbox-150900973.html (2026-07-08)
- Last verified: 2026-10-15

### Run async code-research benchmarks with a coding agent
- URL: https://www.realagentcases.com/use-cases/run-async-code-research-benchmarks-with-a-coding-agent/
- Outcome: An async coding agent cloned several Python Markdown libraries, benchmarked them against cmarkgfm, and wrote a report with charts, with almost no human time spent.
- Agents: Claude Code for web, OpenAI Codex
- Category: Research
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://simonwillison.net/2025/Nov/6/async-code-research/ (2025-11-06)
- Last verified: 2026-10-15

### Compare top social media tools in a spreadsheet with ChatGPT agent and Claude
- URL: https://www.realagentcases.com/use-cases/compare-top-social-media-tools-in-a-spreadsheet-with-chatgpt-agent-and-claude/
- Outcome: Allie K. Miller asked both ChatGPT Agent Mode and Claude's Excel feature to compare top social media tools, and Claude finished in 3 minutes vs 8 with denser decision support, though some prices mixed annual vs monthly.
- Agents: ChatGPT agent, Claude (Excel)
- Category: Research
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/alliekmiller/status/1967631471520211354 (2025-09-15)
- Last verified: 2026-10-15

### Make a UK tax research agent from a folder of HMRC manuals with Claude Code
- URL: https://www.realagentcases.com/use-cases/make-a-uk-tax-research-agent-from-a-folder-of-hmrc-manuals-with-claude-code/
- Outcome: Using scraped HMRC manuals and legislation plus a CLAUDE.md that launches parallel subagents, Claude Code answered an ATT exam question in about 5 minutes, earning roughly 2.5 of 3 marks by the author's grading.
- Agents: Claude Code
- Category: Research
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://martinalderson.com/posts/building-a-tax-agent-with-claude-code/ (2025-08-21)
- Last verified: 2026-10-15

### Build an Nvidia DCF valuation spreadsheet with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/build-an-nvidia-dcf-valuation-spreadsheet-with-chatgpt-agent/
- Outcome: ChatGPT agent ran for 37 minutes, walked through DCF assumptions and returned a downloadable Excel file with formulas and numbers.
- Agents: ChatGPT agent
- Category: Research
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/alliekmiller/status/1946277369360130465 (2025-07-18)
- Last verified: 2026-10-15

### Build an early-retirement (FIRE) plan with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/build-an-early-retirement-fire-plan-with-chatgpt-agent/
- Outcome: With early access, the poster had ChatGPT agent spend 20 minutes researching Vancouver tax law, modeling savings and FIRE scenarios, and producing a downloadable presentation.
- Agents: ChatGPT agent
- Category: Research
- Evidence: Shown on video
- Claimed result: None
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/rowancheung/status/1945896543263080736 (2025-07-17)
- Last verified: 2026-10-15

### Find and summarize files in Google Drive with Comet Assistant
- URL: https://www.realagentcases.com/use-cases/find-and-summarize-files-in-google-drive-with-comet-assistant/
- Outcome: Comet Assistant kept trying until it found the right files in Google Drive and found a way around opening them, but its summary skipped parts of a PRD after reading only about half.
- Agents: Perplexity Comet Assistant
- Category: Research
- Evidence: Shown on video
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Partial. Stopped reading the document about halfway and left out several points from its summary.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/isentropicsoul/status/1944094115819307191 (2025-07-12)
- Last verified: 2026-10-15

### Compile a list of China-tech reporters with Manus
- URL: https://www.realagentcases.com/use-cases/compile-a-list-of-china-tech-reporters-with-manus/
- Outcome: After being pushed to be thorough, Manus produced a downloadable list of 30 journalists covering China tech with their outlets and notable work, though paywalls and captchas got in its way.
- Agents: Manus
- Category: Research
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://www.technologyreview.com/2025/03/11/1113133/manus-ai-review/ (2025-03-11)
- Last verified: 2026-10-15

### Create an interactive elevator-pitch course with Manus
- URL: https://www.realagentcases.com/use-cases/create-an-interactive-elevator-pitch-course-with-manus/
- Outcome: Manus researched the topic and built a complete multi-module course website on elevator pitching; a second prompt added interactive quizzes and videos.
- Agents: Manus
- Category: Research
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.oneusefulthing.org/p/speaking-things-into-existence (2025-03-11)
- Last verified: 2026-10-15

## Shopping

### Clip every digital coupon on a grocery account with Claude Cowork
- URL: https://www.realagentcases.com/use-cases/clip-every-digital-coupon-on-a-grocery-account-with-claude-cowork/
- Outcome: Claude Cowork logged into the poster's Jewel-Osco account and clipped every available coupon.
- Agents: Claude Cowork
- Category: Shopping
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/NYCerInExile/status/2105639156328329221 (2026-10-01)
- Last verified: 2026-10-16

### Prep an Instacart order from history and sales with ChatGPT agent (odd picks)
- URL: https://www.realagentcases.com/use-cases/prep-an-instacart-order-from-history-and-sales-with-chatgpt-agent-odd-picks/
- Outcome: Given access to Instacart and told to prep an order from the poster's history and current sales, ChatGPT agent picked items with reasoning the poster found oddly generic ("cereal, which is on sale and a usual part of groceries").
- Agents: ChatGPT agent
- Category: Shopping
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/osoleve/status/1950032781259477233 (2025-07-29)
- Last verified: 2026-10-16

### Order groceries from H-E-B with ChatGPT agent (access denied)
- URL: https://www.realagentcases.com/use-cases/order-groceries-from-h-e-b-with-chatgpt-agent-access-denied/
- Outcome: The poster tried to order groceries from H-E-B in Austin with ChatGPT agent, but it failed because access was denied.
- Agents: ChatGPT agent
- Category: Shopping
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Failure. Access to the store site was denied, so no order was placed.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/CodyBaker_xx/status/1948746246341943795 (2025-07-25)
- Last verified: 2026-10-16

### Turn 200 photos of household items into a moving-sale list with Claude Code
- URL: https://www.realagentcases.com/use-cases/turn-200-photos-of-household-items-into-a-moving-sale-list-with-claude-code/
- Outcome: Claude Code processed over 200 photos of the poster's household items and created a moving-sale list with descriptions and prices; the poster says it saved hours compared with writing Facebook Marketplace listings by hand.
- Agents: Claude Code
- Category: Shopping
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/rautility/status/1937172819215065456 (2025-06-23)
- Last verified: 2026-10-16

### Compare prices for 60–70 garden plants across 5 sites with OpenAI Operator
- URL: https://www.realagentcases.com/use-cases/compare-prices-for-60-70-garden-plants-across-5-sites-with-openai-operator/
- Outcome: Overnight, Operator cross-referenced a list of 60–70 plants across 5 sites into a price comparison table; the poster called it buggy but fun.
- Agents: OpenAI Operator
- Category: Shopping
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/b1rdmania/status/1931618971968544834 (2025-06-08)
- Last verified: 2026-10-16

### Find working coupon codes at checkout with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/find-working-coupon-codes-at-checkout-with-chatgpt-agent/
- Outcome: ChatGPT agent kept trying coupon codes until it found ones that worked; the poster says it saved them $79.
- Agents: ChatGPT agent
- Category: Shopping
- Evidence: Screenshots or logs
- Claimed result: $79 saved (Poster says; a claim from the source, not independently verified)
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/joevkolo/status/1976496831623463109 (2025-10-10)
- Last verified: 2026-10-15

### Find local takeout deals under a budget with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/find-local-takeout-deals-under-a-budget-with-chatgpt-agent/
- Outcome: The Tom's Guide author uses ChatGPT agent to scour delivery apps and local restaurants for family takeout deals under a set budget.
- Agents: ChatGPT agent
- Category: Shopping
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.tomsguide.com/ai/5-clever-ways-i-use-chatgpt-agent-to-save-serious-money-each-month (2025-10-10)
- Last verified: 2026-10-15

### Daily product restock check with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/daily-product-restock-check-with-chatgpt-agent/
- Outcome: A scheduled ChatGPT agent checks every day whether a product is in stock, reports at 8am, and even flags status changes like 'out of stock' becoming 'coming soon'.
- Agents: ChatGPT agent
- Category: Shopping
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/starsailing11/status/1969125311565430944 (2025-09-19)
- Last verified: 2026-10-15

### Order weekly meal delivery with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/order-weekly-meal-delivery-with-chatgpt-agent/
- Outcome: ChatGPT agent logged into Gym Meals Direct, chose meals and sizes, picked Monday delivery and applied a discount.
- Agents: ChatGPT agent
- Category: Shopping
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/benny_alexander/status/1952206488023585198 (2025-08-04)
- Last verified: 2026-10-15

### Draft and complete an eBay Germany listing from photos with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/draft-and-complete-an-ebay-germany-listing-from-photos-with-chatgpt-agent/
- Outcome: From a few photos, ChatGPT agent identified an Apple Watch Series 7, noted display scratches, drafted a German title and description, then finished the eBay DE listing draft after login in about 15 minutes.
- Agents: ChatGPT agent
- Category: Shopping
- Evidence: Shown on video
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/rschu/status/1950470327252705510 (2025-07-30)
- Last verified: 2026-10-15

### Request a furniture quote with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/request-a-furniture-quote-with-chatgpt-agent/
- Outcome: In 11 minutes ChatGPT agent ran 13 searches, skipped JavaScript-heavy sites it couldn't read, and shortlisted wicker daybeds from server-rendered retailers so it could request a quote in Sydney.
- Agents: ChatGPT agent
- Category: Shopping
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.leannewong.co/chatgpt-agent-ai-agentic-seo/ (2025-07-24)
- Last verified: 2026-10-15

### Fill an online grocery cart from a handwritten list with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/fill-an-online-grocery-cart-from-a-handwritten-list-with-chatgpt-agent/
- Outcome: ChatGPT agent transcribed 15 of 16 items from a photographed shopping list (missing onions), found the nearest Harris Teeter and added them to the cart; a safety monitor first blocked the Kroger login, but after the user said they'd log in themselves and entered the password, it ordered the groceries.
- Agents: ChatGPT agent
- Category: Shopping
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Partial. Missed one of 16 items and gave up at the store login until the user stepped in; the author called it still not very useful.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://www.understandingai.org/p/chatgpt-agent-a-big-improvement-but (2025-07-22)
- Last verified: 2026-10-15

### Order groceries from a 16-item list with OpenAI Operator
- URL: https://www.realagentcases.com/use-cases/order-groceries-from-a-16-item-list-with-openai-operator/
- Outcome: Operator misread parts of a 16-item grocery list and added only 13 items to the cart even after corrections; the author concluded it wasn't ready for real use.
- Agents: OpenAI Operator
- Category: Shopping
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://www.understandingai.org/p/computer-use-agents-seem-like-a-dead (2025-06-26)
- Last verified: 2026-10-15

## Travel

### Find flights with Manus (16-hour run, poor pick)
- URL: https://www.realagentcases.com/use-cases/find-flights-with-manus-16-hour-run-poor-pick/
- Outcome: Asked to find flights, Manus ran for 16 hours and then picked what the poster called probably the most expensive option, with a 9-hour layover.
- Agents: Manus
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Failure. Ran for 16 hours, then picked what the poster called probably the most expensive option, with a 9-hour layover.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/natalienkhalil/status/2107519229109182513 (2026-10-06)
- Last verified: 2026-10-16

### Plan a six-week Southeast Asia trip with Manus (hallucinated flights)
- URL: https://www.realagentcases.com/use-cases/plan-a-six-week-southeast-asia-trip-with-manus-hallucinated-flights/
- Outcome: The poster used Manus to plan a 6-week Southeast Asia trip, and it hallucinated flights that didn't exist despite having browser access.
- Agents: Manus
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Failure. Recommended flights that did not exist, despite having browser access.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/MattMickiewicz/status/2103222377459188214 (2026-09-24)
- Last verified: 2026-10-16

### Book every restaurant for a Japan trip with Muse
- URL: https://www.realagentcases.com/use-cases/book-every-restaurant-for-a-japan-trip-with-muse/
- Outcome: Muse booked every restaurant for the poster's Japan trip and corresponded in Japanese with a Hakone hotel to secure a dinner slot not offered on its website.
- Agents: Muse
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/Han_Fang_/status/2101874865385169348 (2026-09-21)
- Last verified: 2026-10-16

### Book a whole weekend trip in 30 minutes with ChatGPT Work
- URL: https://www.realagentcases.com/use-cases/book-a-whole-weekend-trip-in-30-minutes-with-chatgpt-work/
- Outcome: ChatGPT read the poster's chat with their girlfriend, proposed itinerary options, staged an Airbnb, rental car, fishing gear and dinner reservations through browser use, then made a shareable trip page after the poster booked.
- Agents: ChatGPT Work
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/nickbaumann_/status/2087598230968373584 (2026-08-12)
- Last verified: 2026-10-16

### Build a work-trip itinerary from booking emails with Claude
- URL: https://www.realagentcases.com/use-cases/build-a-work-trip-itinerary-from-booking-emails-with-claude/
- Outcome: The day before a work trip, Claude read the poster's booking emails and organized a polished itinerary that the poster says would have taken a few hours by hand.
- Agents: Claude
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/andreistenmark/status/2064466188764418315 (2026-06-09)
- Last verified: 2026-10-16

### Plan a family Disneyland trip with Claude Code and Skills
- URL: https://www.realagentcases.com/use-cases/plan-a-family-disneyland-trip-with-claude-code-and-skills/
- Outcome: The poster planned a family Disneyland trip using Claude Code as the agent, Claude Skills for research, planning, tracking and journaling, and a simple companion web app for itineraries and maps.
- Agents: Claude Code
- Category: Travel
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/gautam/status/2012601709000814926 (2026-01-17)
- Last verified: 2026-10-16

### Plan a points-funded family holiday and book dinner with ChatGPT agent (partial)
- URL: https://www.realagentcases.com/use-cases/plan-a-points-funded-family-holiday-and-book-dinner-with-chatgpt-agent-partial/
- Outcome: For a points-funded family holiday plan, ChatGPT agent built a decent deck but spent 15 minutes formatting slides; it couldn't book the romantic dinner.
- Agents: ChatGPT agent
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Partial. Built a decent deck but spent 15 minutes formatting slides and could not book the dinner.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/hotelemarketer/status/1949081438076244273 (2025-07-26)
- Last verified: 2026-10-16

### Check a hotel price with ChatGPT agent (wrong town)
- URL: https://www.realagentcases.com/use-cases/check-a-hotel-price-with-chatgpt-agent-wrong-town/
- Outcome: Asked to check a hotel's price with its town clearly stated, ChatGPT agent looked up a same-named hotel in another town, and the poster notes this wouldn't have been obvious from the reply.
- Agents: ChatGPT agent
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Failure. Looked up a same-named hotel in the wrong town, and the reply did not make that obvious.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/szilagyipal/status/1948670475959501318 (2025-07-25)
- Last verified: 2026-10-16

### File Thailand Digital Arrival Cards for five travelers with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/file-thailand-digital-arrival-cards-for-five-travelers-with-chatgpt-agent/
- Outcome: ChatGPT agent filled Thailand Digital Arrival Cards for 5 people (about an hour / 16 credits); the poster verified the preview and swapped in real passport numbers before submitting.
- Agents: ChatGPT agent
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/SatanAtWink/status/2006301585996169399 (2025-12-31)
- Last verified: 2026-10-15

### Hunt last-minute holiday flights with multiple agents (failure case)
- URL: https://www.realagentcases.com/use-cases/hunt-last-minute-holiday-flights-with-multiple-agents-failure-case/
- Outcome: Claude Opus 4.5, Gemini 3 Pro and ChatGPT agent mode only matched a manual Google Flights search and were too slow as cheap fares vanished; the humans beat them by about 15%.
- Agents: ChatGPT agent, Claude Opus 4.5 (Antigravity), Gemini 3 Pro
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Failure. The agents only matched a manual Google Flights search and were too slow as cheap fares vanished; the humans beat them by about 15%.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/anuraagsidhu/status/2001930515260088374 (2025-12-19)
- Last verified: 2026-10-15

### Compare browser agents on a vacation Airbnb search
- URL: https://www.realagentcases.com/use-cases/compare-browser-agents-on-a-vacation-airbnb-search/
- Outcome: On the same Airbnb search, ChatGPT Atlas agent mode and Comet each found properties in about 5 minutes, Claude's extension took 10 minutes to find the first, and MiniMax Agent ran over an hour without giving links.
- Agents: ChatGPT Atlas agent mode, Perplexity Comet, Claude in Chrome, MiniMax Agent
- Category: Travel
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/oyacaro/status/1982805076952465599 (2025-10-27)
- Last verified: 2026-10-15

### Plan St. Simons Island restaurants with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/plan-st-simons-island-restaurants-with-chatgpt-agent/
- Outcome: ChatGPT Agent Mode produced a dining guide for a 4-day St. Simons Island trip organized by meal and price, and the author ate at 4 restaurants from the list and liked them all.
- Agents: ChatGPT agent
- Category: Travel
- Evidence: Article or interview
- Claimed result: None
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://www.theaienterprise.io/p/ai-agents-research (2025-10-21)
- Last verified: 2026-10-15

### Search for cheap flights with ChatGPT agent (failure case)
- URL: https://www.realagentcases.com/use-cases/search-for-cheap-flights-with-chatgpt-agent-failure-case/
- Outcome: ChatGPT agent suggested flights costing $1,400 when the poster found an $850 option on their own.
- Agents: ChatGPT agent
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Failure. Suggested a $1,400 flight; the poster found an $850 option on their own.
- Prompt type: Improved (strengthened by us to avoid the problem in the source; not tested by the poster). What the poster tried: recreated by us
- Source: https://x.com/maxdrescher_/status/1958253813770674556 (2025-08-20)
- Last verified: 2026-10-15

### Find the cheapest one-way flight across two departure cities with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/find-the-cheapest-one-way-flight-across-two-departure-cities-with-chatgpt-agent/
- Outcome: From a vague brief, ChatGPT agent searched flights from Ljubljana or Zagreb to Bordeaux over a two-week window and found a €141.48 fare in 12 minutes.
- Agents: ChatGPT agent
- Category: Travel
- Evidence: Screenshots or logs
- Claimed result: None
- Beginner-friendly: Yes
- Outcome status: Worked (per the source)
- Prompt type: Original (as published in the source)
- Source: https://x.com/GasperCrepinsek/status/1951266026324205864 (2025-08-01)
- Last verified: 2026-10-15

### Plan a Budapest weekend and start an Airbnb booking with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/plan-a-budapest-weekend-and-start-an-airbnb-booking-with-chatgpt-agent/
- Outcome: ChatGPT agent researched accommodation, activities and vegan restaurants for a Budapest weekend and got into an Airbnb reservation flow, then handed over to the user for login and payment.
- Agents: ChatGPT agent
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://www.visibletourism.com/post/from-zero-click-to-zero-effort-chatgpt-agent-mode/ (2025-07-25)
- Last verified: 2026-10-15

### Book an Airbnb with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/book-an-airbnb-with-chatgpt-agent/
- Outcome: ChatGPT agent booked what the poster called the perfect Airbnb on the first try.
- Agents: ChatGPT agent
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/romainhuet/status/1946015036503449813 (2025-07-18)
- Last verified: 2026-10-15

### Find a family vacation rental in Costa Rica with ChatGPT agent
- URL: https://www.realagentcases.com/use-cases/find-a-family-vacation-rental-in-costa-rica-with-chatgpt-agent/
- Outcome: ChatGPT agent found a Costa Rica place with a pool at a decent rate for 2 adults and 3 kids; the poster was tempted to book without checking further.
- Agents: ChatGPT agent
- Category: Travel
- Evidence: Described by poster
- Claimed result: None
- Beginner-friendly: No
- Outcome status: Worked (per the source)
- Prompt type: Recreated (the source did not publish the exact prompt)
- Source: https://x.com/kalenjordan/status/1946285767690727918 (2025-07-18)
- Last verified: 2026-10-15