How to use Claude to turn web apps into MCP tools
A captured HAR file became OpenAPI documentation, followed by a local Python MCP server that used Playwright to obtain credentials before calling internal APIs.
Added1125 use casesShowing 217–240
Preparing search…
A captured HAR file became OpenAPI documentation, followed by a local Python MCP server that used Playwright to obtain credentials before calling internal APIs.
AddedClaude completed overdue upgrades on a branch while the author checked in occasionally. The team then performed several engineer days of manual testing and deployed.
AddedAn author used Claude Code to clean up a routing app built with Antigravity, then live-tested its shadow-aware routes and identified missing tree data.
AddedClaude Code helped refactor a C# reporting DLL, add dependency injection and tests, and create a CLI; the developer repaired logic errors before final release testing.
AddedA React developer reviewed an initial Claude Code refactor before expanding it to hundreds of edits; most worked, but declaration errors required manual repairs.
AddedGemini Pro 3 generated illumination and camera drivers before rewriting a Python Qt microscope stack; the developer guided the work and tested the replacement.
AddedClaude Code drafted image collection, labeling and PyTorch prediction tools for a subway-status app, but the author needed substantial manual fixes.
AddedClaude Code coordinates implementation lanes, review and CI before a lander merges and deploys; the author handles decisions and periodically checks the product.
AddedClaude Code helped restore a PRQL PostgreSQL extension using upstream source inspection and existing tests; the maintainer then manually refactored it.
AddedOssature’s author used specification audits and scoped build-plan tasks to build a CHIP-8 emulator, saving prompts and responses rather than relying on chat history.
Added OssatureCodex extracted and analyzed pi sessions to shape an outer guidance prompt; the author reported mixed results and reduced check-in time on a game.
AddedMillrace runs a governed loop that ports Python releases to Rust, checks parity and publishes validated versions; the underlying agent is unnamed.
Added MillraceCodex refactored a tower defense game and reported matching screenshot hashes after correctly recapturing the baseline, following screenshot instructions in PLANS.md.
AddedAn agent-swarm contributor used AI to generate TLA+ specifications and explain them, then turned counterexamples into failing tests and fixes.
Added Unspecified AI agentAn engineer used an agent-built Azure DevOps CLI to combine cards, meeting transcripts, requirements, and repository context before approving batch updates.
Added Unspecified coding agentA solo developer used separate Cursor architect and programmer windows, with written instructions preceding implementation, and reported less architectural drift.
AddedPloy corrected evaluation assumptions, reran homepage redesign comparisons, and examined tool-call failures before adopting GPT-5.6 Sol for its website agent.
Added Ploy agentA founder developed Linkeme.ai from LinkedIn automation into a multi-platform publishing product, then migrated its workflows to AWS using Cline.
Added ClineThe author reports a Rust/Bevy sandbox running into the thousands of simulation agents. A 100,000-agent target remained unachieved, and the sandbox had no game objective.
AddedThe report lists $3.12 per manual transaction versus $0.17 per batched transaction. The author cautions that realistic savings may be around 10–12x rather than the table’s 18.6x.
AddedSafronov reported an instant tax-file download at the library. Engye grew to support sending and collecting files, clipboard sharing and a virtual drive.
AddedThe author reports finding an Avis reseller discount and describes $1,000 in savings for both cars. The quoted $600 versus $1,100 pricing is ambiguous about the booking total.
AddedLin reported a peak of approximately 2,000 concurrent agents. FastRender displayed GitHub, Wikipedia and CNN pages, but JavaScript remained disabled and the project was a research experiment.
AddedThe engineer reports 1,467+ replies automatically sent to real buyers and unattended operation of a chat channel without an official API.
Added Custom RAG customer-service agent