Claude for Open Source · Application read

Six months of free Claude Max. Here is our honest shot at it.

Anthropic is giving 6 months of Claude Max 20x to open-source maintainers. This is the real picture: where we actually stand against their rules, what we can truthfully point to, and exactly what you would submit if you want to take the swing.

Verdict · Long shot

We clear none of the hard bars. The only honest route is the "apply anyway" clause.

The program has five numeric tests (downloads, dependents, merged pull requests, contributors, criticality score). We meet zero of them. There is a catch-all for people who "maintain something the ecosystem quietly depends on," and that is the one honest door open to us.

Odds are low. What is genuinely strong is one thing: real, recent, verifiable shipping activity by an independent builder. That is worth an honest application. It is not worth padding, because every number here is something Anthropic can check in thirty seconds.

The offer

What is on the table

6 mo

Claude Max 20x

The ~$200/month tier, free for the period. No auto-renew; reverts to your prior plan after.

10,000

Recipient cap

Rolling review. Activate within 90 days of approval or the benefit is forfeited.

One subscription per person, non-transferable, base plan only (overages billed normally). Applying does not commit you to anything.

Our footprint · verified live

What we can truthfully point to

TreeTrace GitHub stars

Public, Apache-2.0, created 2026-06-15. 0 forks. Small and new, stated plainly.

3,438

npm installs to date

Package treetrace, essentially its whole life so far. 63 last week.

Releases shipped

v0.4.0 to v0.10.4, latest 2026-07-05. All authored by zionboggan. This is the real strength: cadence.

Other active repos

oversight, memory-graph, docket and more on the same account, all pushed in the last 3 months (all at 0 stars).

Recent activity is genuine and easy to verify. That criterion we actually pass. The problem is reach, not recency.
A methods write-up backs the story at zionboggan.com/research.html, on shipping a routed crew of agents. Supporting narrative, not a credential.

The rules vs. us

Scorecard against the live criteria

Their bar	What it takes	Where we are	Meets
Library author	500+ dependent repos, or 100+ dependent packages, or 200k+ monthly downloads	~3,438 total installs, 0 dependents	No
Core contributor	Listed committer on CPython, Rust, Node, Apache, CNCF and similar	None	No
Active contributor	100+ pull requests merged into repos you do not own, last 12 months	Not close	No
Community builder	A repo with 20+ external contributors (merged PRs)	Commits are solo, 0 forks	No
Critical infra	OpenSSF criticality score 0.4 or above	A 1-month, 3-star repo scores well below	No
Catch-all	"Something the ecosystem quietly depends on. Apply anyway and tell us."	Real, active, Apache tool. Early, but honest.	Our door

One thing to be careful with

The Flowise security find

Do not cite it as a CVE yet

Our notes reference a credited advisory (GHSA-rf9q-wwm3-q36v) on FlowiseAI, a 54,000-star project. As of today that advisory ID does not resolve publicly: it 404s on GitHub and returns nothing in advisory search. It may still be private or under triage.

Anthropic can check a claimed CVE instantly. A credit a reviewer cannot find hurts more than it helps. Word it softly if you include it ("reported an issue, maintainers acknowledged it, advisory not yet public"), and only upgrade to a hard citation once it is published and the exact ID is confirmed.

If you want to take the swing

Exactly what you submit

Open the official pageGo to claude.com/contact-sales/claude-for-oss and click Apply now. The form is only visible once you click, so I could not pre-fill the exact fields.
Apply under the catch-allSay plainly you do not hit a numeric bar and are applying as an active independent maintainer. Honesty is the whole strategy here.
Point to TreeTraceRepo github.com/TreeTraceTool/TreeTrace, npm treetrace, 9 releases, most recent 2026-07-05. Role: sole maintainer.
Add the wider body of workYour github.com/zionboggan profile and the research write-up at zionboggan.com/research.html.
Send it as yourselfThe full text is ready in APPLICATION-DRAFT.md. It is your account and your call. I have not submitted anything.

The draft is written and waiting

Every answer is filled in with real, checkable numbers and worded with no AI tells. Review it, decide if the long shot is worth your one attempt, and if so, paste it into the live form yourself.

Open the program page →