Show HN: Machine Internet: turn a site's hidden JSON API into MCP tools
Hello HN, I'm Chetan, a CS student at the University of Alberta. There are many sites that have no official API, but their frontend makes calls to a perfectly good JSON API. Machine Internet discovers it and converts it to MCP tools that your agent can use. uvx machine-internet wrap <url> It loads the site in a headless browser, logs the JSON requests the site makes, and an LLM converts those into tools. Unlike tools that scrape page content, your agent calls the live API directly. It work
Read original sourceSource record
Recorded metrics
points · 2 votes
2026-11-17 18:16:14.395Z
cumulative since publication
Research analysis
Project record
Repository: chetanty/machine_internet. Source account: Hello HN, I'm Chetan, a CS student at the University of Alberta. There are many sites that have no official API, but their frontend makes calls to a perfectly good JSON API. Machine Internet discovers it and converts it to MCP tools that your agent can use. uvx machine-internet wrap <url> It loads the site in a headless browser, logs the JSON requests the site makes, and an LLM converts those into tools. Unlike tools that scrape page content, your agent calls the live API directly. It work
Observed adoption & comparisons
Source votes: 2 votes (Nov 17, 2026 UTC; cumulative since publication) Recent records to compare: “affaan-m/ECC” (Trendshift, Nov 19, 2026); “mvanhorn/last30days-skill” (Trendshift, Nov 19, 2026). Shared themes are navigation hints; they have not been verified as the same need or independent corroboration.
Adoption decision
Evaluate it against the tool or manual workflow it would replace. The stored record supports discovery; it does not yet establish reliability, lower operating cost or commercial adoption.
Unverified requirements
Not verified: license permissions, required hardware, installation steps, maintenance status and a representative task with its failure cases.
Next test
Choose one real task and a current alternative. Record the commit, environment, input, output, elapsed time and failure behavior; retain the result before recommending adoption.