# localtunnel

> localtunnel.github.io | Egress and tunnels; 1 record: free tier, limits, identity cost and evidence.

HTML https://fshy.ajam.dev/p/localtunnel-github-io/ | JSON https://fshy.ajam.dev/api/v1/entity/localtunnel-github-io.json | how to read a record: https://fshy.ajam.dev/llms-full.txt

## localtunnel

- `egress/localtunnel` | qualifies-with-caveat | evidence documented | binding none | category tunnel (hosted, shared instance)
- needs card no, phone no, ID no, institution no | verified 2026-11-03T10:32:57.641Z (live) | stale on 2026-12-04
- url https://localtunnel.github.io/www/
- free tier: a unique public URL proxied to a local server, with no published limits | unit: unlimited | metered on: request-count | read on: 2026-10-22 | source: [1]
- per-object cap: not applicable: a stream
- retention: the assigned subdomain is random per run | source: [1]
- egress: no quota published anywhere: the project page states a URL and no limit | metered: no | source: [1]
- identity: anonymous | detail: no account to use the shared server
- interface: cli
- limits:
  - per-ip-shaping: no published limit: the original author wrote 'There are no limitations - abuse happens, server goes down, comes back up sometime later' [2]
- binding note: No byte figure exists anywhere: the author says there are no limitations and the current maintainer explains outages as CPU saturation, so the failure mode is the single volunteer instance rather than a meter. A user measured about 0.8 Mbit/s download through it in 2019, which would cap a month near 260 GB, but that is that instance on that day.
- interface detail: npx localtunnel --port PORT against the shared localtunnel.me server
- tooling: the npm package, or the standalone binary
- composes with: composes with an origin and nothing that needs a published ceiling; the shared server is someone else's capacity and is the first thing to break.
- reliability: the shared server is a single volunteer-run instance with no published capacity
- notes: Not measured: the shared server is one instance and a throughput number from it would describe that instance on that day. The record carried an empty limits array. Replaces 'nothing binds' with the reason no byte constraint exists and the one throughput datum that stands in for it.
- extras:
  - stale_after_days: 180
  - bytes_transit: yes
  - transit_note: the shared server proxies the origin's bytes
- evidence:
  - [1] 2026-10-22: the project page states that it assigns a unique public URL and says nothing about a limit, which is a documented absence
  - probe:https://localtunnel.me/ 2026-10-22: the bare host answers 302 and the project docs live at localtunnel.github.io/www/, so there is no vendor pricing page to read
  - [2] 2026-11-03: a project contributor states there is no published limit on the shared server | capture: sha256:0f9481d7e29e9e13e49d3666c4ef7ba94803870678877db4435e12d79efafb6b | quote: There are no limitations - abuse happens, server goes down, comes back up sometime later :)
  - [1] 2026-11-03: the project page states a public URL proxied to a local server and lists no byte or rate limit | capture: sha256:e06dfea03a86ad96d64a06912b48b36a7067d35311a47ac16fd132f080c813f9 | quote: Localtunnel will assign you a unique publicly accessible url that will proxy all requests to your locally running webserver.
- sources:
  - [1] https://localtunnel.github.io/www/
  - [2] https://github.com/localtunnel/localtunnel/issues/288