The issue asked to confirm whether email is ever verified. It is: the
feature is built, and it gates issuing a TOKEN rather than signing up —
which is exactly the 'verify to send, not to sign up' trade the issue
proposed.
Measured on production: all 6 accounts arrive via the shared FlameNet
service, ZERO have a local email, and the relay has no SMTP configured at
all. So the exposure described — a reset path aimed at an address nobody
owns — does not exist here. It is a self-hosting concern, and for
self-hosters the feature already exists behind
FN_EMAIL_VERIFICATION_REQUIRED.
The one real gap was where it was written down. docs/CONFIGURATION.md
describes it accurately; .env.example did not mention it, and .env.example is where somebody configuring SMTP actually looks —
which is the same person who has just become able to turn it on.
Placed inside the SMTP block on purpose: enforcement is required AND canSendMail, so setting it on a relay with no mail does
nothing. That is deliberate — an account that cannot be sent a code could
never spend one — but it is the kind of silent no-op somebody should meet
in the comment rather than in production.
The issue asked to confirm whether email is ever verified. It is: the
feature is built, and it gates issuing a TOKEN rather than signing up —
which is exactly the 'verify to send, not to sign up' trade the issue
proposed.
Measured on production: all 6 accounts arrive via the shared FlameNet
service, ZERO have a local email, and the relay has no SMTP configured at
all. So the exposure described — a reset path aimed at an address nobody
owns — does not exist here. It is a self-hosting concern, and for
self-hosters the feature already exists behind
FN_EMAIL_VERIFICATION_REQUIRED.
The one real gap was where it was written down. `docs/CONFIGURATION.md`
describes it accurately; `.env.example` did not mention it, and
`.env.example` is where somebody configuring SMTP actually looks —
which is the same person who has just become able to turn it on.
Placed inside the SMTP block on purpose: enforcement is
`required AND canSendMail`, so setting it on a relay with no mail does
nothing. That is deliberate — an account that cannot be sent a code could
never spend one — but it is the kind of silent no-op somebody should meet
in the comment rather than in production.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
The issue asked to confirm whether email is ever verified. It is: the
feature is built, and it gates issuing a TOKEN rather than signing up —
which is exactly the 'verify to send, not to sign up' trade the issue
proposed.
Measured on production: all 6 accounts arrive via the shared FlameNet
service, ZERO have a local email, and the relay has no SMTP configured at
all. So the exposure described — a reset path aimed at an address nobody
owns — does not exist here. It is a self-hosting concern, and for
self-hosters the feature already exists behind
FN_EMAIL_VERIFICATION_REQUIRED.
The one real gap was where it was written down.
docs/CONFIGURATION.mddescribes it accurately;
.env.exampledid not mention it, and.env.exampleis where somebody configuring SMTP actually looks —which is the same person who has just become able to turn it on.
Placed inside the SMTP block on purpose: enforcement is
required AND canSendMail, so setting it on a relay with no mail doesnothing. That is deliberate — an account that cannot be sent a code could
never spend one — but it is the kind of silent no-op somebody should meet
in the comment rather than in production.
Co-Authored-By: Claude Opus 5 noreply@anthropic.com
assigned to @paulhitt
mentioned in issue #40
mentioned in commit
07408fd400changed the description