Product1 distinct publisher3 min readUpdated
Snowflake Volunteer ships on Google Play and F-Droid, built with Bloco and Guardian Project. The pool it feeds sat near 1,700 daily proxies, still short of a 2,100 peak.
The Product Desk · Product desk

Compiled by The Product DeskSomething wrong?How this is made
The Tor Project and its partners have shipped Snowflake as a standalone Android app, Snowflake Volunteer, on the Google Play Store, where anyone can install it and lend their connection as a proxy into the Tor network [1][3]. That moves proxy running from a task for people with spare bandwidth and the patience to configure a node into something closer to a one-tap install, and it makes the size of Tor's bridge pool partly a function of consumer app distribution [9][14][10].
The mechanics have not changed. Snowflake, which debuted in 2019, dresses a client's traffic up as a WebRTC video call and routes it through short-lived proxy connections that keep moving, which is harder for a censor to enumerate than a static server address [5][7]. When one proxy is blocked, the client jumps to another [8]. The pool is the product, so headcount is the metric. Neither Snowflake nor the new app is meant for people inside censored countries; both are for volunteers in places where Tor is still reachable [6].
The numbers are the interesting part, and they are modest. ZDNET reports that Tor counted roughly 1,300 unique proxy IP addresses active per day by May, after an April launch, rising to about 1,700 a month later, which the piece describes as a 29% month-over-month increase [11][12]. On the rounded figures given, the jump is 400 proxies, or about 31% [15]. The same report puts the historical peak above 2,100 daily active proxies, so the pool at 1,700 is running roughly 19% below its own high-water mark [13][16]. A distribution channel that reaches ordinary Android users is the plausible way to close that gap, and the app is deliberately narrow to make that easy: the developers call it a single-purpose app that gives volunteers control over when and how they contribute [4], and ZDNET's hands-on notes it surfaces a volunteer's contribution in real time [17][19].
The trade is a new dependency. A Play Store listing is one object in one company's catalogue, and a proxy network whose recruitment funnel runs through it inherits whatever happens to that listing. The hedge is already in place: Snowflake Volunteer is also on F-Droid, the code is open source, and developers can inspect it on GitHub or fork it [10]. Nothing in the source material says any government has moved against the listing, and we are not claiming one has. The structural point stands regardless, because the reason Snowflake is hard to block is that it has no fixed address to attack [7], while its main recruitment surface now does.
Worth watching: whether daily active proxies clear the 2,100 peak now that installation does not require technical knowledge [13][9], and whether growth from Play Store installs holds or decays as consumer apps usually do after a launch spike. Also worth watching is the split between Play and F-Droid installs, since that ratio is the practical measure of how much of the pool would survive a single store removal [10].
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
Snowflake Volunteer, a new Android app from the people behind the Tor Project, has landed on the Google Play Store as a way for the general public to help fight global censorship.
Snowflake Volunteer is a standalone Android app that allows users to lend their internet connections to others to create Snowflake proxies for accessing the Tor network; the user's mobile device becomes one of these proxies.
Snowflake Volunteer is described as a "single-purpose app that gives volunteers control over when and how they contribute."
Neither Snowflake nor Snowflake Volunteer is designed for people living in an oppressed country; they are projects for individuals living in countries that do allow access to the Tor network and are willing to lend that access to others.
The app turns a phone into a proxy and requires no technical knowledge.
Snowflake Volunteer can be downloaded on Google Play and F-Droid; its code is open source, and developers can explore the app on GitHub or create their own forks.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Single-publisher hands-on plus self-reported pool metrics
Distribution and app behavior are directly observed by the reviewer, which is reasonably firm evidence for the product's existence and controls. Every quantitative claim, however, comes from one outlet relaying figures attributed to Tor, with no independent measurement, no dated metric series, and no corroborating publisher in the cluster. The internal 29% vs ~31% inconsistency further limits how much weight the numbers can carry.
Real distribution, modest and still-below-peak proxy pool
Adoption is genuinely observable rather than promised: the app is live on two app stores and the proxy pool it feeds has a reported trajectory from ~1,300 to ~1,700 daily proxies. The absolute scale is small for a global censorship-circumvention network, the level remains roughly 19% under the reported 2,100 peak, and no install counts, active-volunteer counts, or bytes-served figures are given, so the score stays in the low band.
Mildly overstated by consumer-review framing
The framing that 'anyone can fight censorship' and that the app 'shows your impact in real time' runs ahead of what the evidence establishes: a pool of a few thousand daily proxies that is still below its own peak, with a growth rate the article itself states imprecisely. The underlying facts are real and modestly evidenced, so the gap is a moderate positive rather than a large one.
Advocacy metrics relayed through a consumer review
The only quantitative claims originate with the party that benefits from more volunteers, and the article repeats the developers' own positioning language, including their argument that help is 'already in our pockets' and the 'single-purpose app' description. Offsetting factors are visible in the material itself: the software is free, open source, and forkable, with no paid tier or vendor lock-in disclosed, so the pull is mission-recruitment rather than revenue.
Product facts solid, numbers thinly sourced
Confidence is limited by cluster breadth: one publisher, one hands-on account, and self-reported metrics with an internal arithmetic inconsistency. The product-level claims (distribution channels, partnership, controls, WebRTC mechanism) are consistent and specific enough to rely on; the adoption trajectory should be treated as indicative only.
product
Most AI-justified layoffs cost more than they saved, and half the cutters may rehire by 20271 distinct publisher
build
Sideloading becomes a registered activity: budgeting for Android's September 2026 deadline1 distinct publisher
build
A cleanup commit deleted the sanitizer. Five days later a scanner cashed it in.1 distinct publisher
product
The AI-wrote-it claim died in eight hours. The Actions injection pattern did not.1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 14, 2026