| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- pay_keysend never called checkBudget and never recorded to
${pubkey}:payments, so a budget-limited connection string was a
full-drain credential. Budget-check it, honor max_fee, and record the
debit so subsequent budgeted calls count it (S4).
- Replace the non-atomic zScore-then-(unawaited)-zAdd dedup with an
awaited SET NX claim, so two concurrent deliveries of the same signed
event can't both reach handle() and pay twice; also reject
future-dated created_at beyond a small skew (S6).
- Serialize budget-affecting methods per app with an async mutex so a
concurrent request can't read a stale spent-total and overspend (S8).
Co-Authored-By: Claude Opus 4.8 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
| |
POST /upload/:type fed the raw uploaded image straight into sharp with
no auth preValidation (unlike neighboring routes), reaching a native
libvips decoder in the container that holds the JWT secret and CLN hot
wallet. Require auth, and bump sharp 0.33.5 -> 0.35.3 (out of the
vulnerable < 0.35.0 libvips range, GHSA-f88m-g3jw-g9cj et al.).
Co-Authored-By: Claude Opus 4.8 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
| |
The request logger persisted full req.headers to the "req" log,
including the Authorization bearer token and the token cookie — both
non-expiring JWTs, so log-read access became account takeover. Redact
the authorization and cookie headers. Also set httpOnly/secure/sameSite
on the token cookie so it isn't readable by injected JS or sent over
plaintext HTTP. (Revocation is already handled by the evicted set.)
Co-Authored-By: Claude Opus 4.8 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The master-password path (isAdmin = password === config.adminpass) logs
in as any user, skipping password, 2FA, and captcha. Worse, when
adminpass is unset an omitted password field made undefined === undefined
true, authenticating as anyone. Require both the configured value and the
supplied password to be non-empty and match exactly, across every
adminpass comparison (login, recaptcha bypass, support email, freeze).
Also scrub config.ts.sample: the sample shipped a working-looking JWT
secret, two server nsec keys, and adminpass "pwa". Replace with
fail-closed placeholders and regenerate instructions. Rotate any
deployment that ever shipped the sample values.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
| |
Registration, password change, and admin reset all hashed at bcrypt
cost 4 (16 rounds, ~1,300 guesses/s/thread) — the single factor gating
the custodial balance and the nip49 passphrase for the stored nostr key.
Raise all three sites to cost 12, and on successful login re-hash any
legacy cost-4 hash now that we hold the verified plaintext.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
The websocket auth gate used jwt.decode(), which parses the payload but
never checks the signature — anyone could forge {id: <victim uid>}
(alg:none or any key), register their socket under the victim's uid, and
receive every emit for that account. Switch both track() and the "login"
gate to jwt.verify(token, config.jwt), rejecting forged/altered tokens.
Also project the full-record socket emits through pick(user, whitelist)
so a "user" event can never carry the bcrypt hash, TOTP secret, or PIN —
matching the shape /login and /me already return.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Serving local accounts unconditionally sent migrated names (whose legacy
user record still exists) back to the old account, and left v3-only
custom names 500ing — only npub1* names had a Cloudflare redirect.
Resolution order is now: registrar-claimed -> registrar (response handed
through, cached 60s, fail-open to local if the registrar is down), else
local account, else 404. The migrated flag stays out of it: register.ts
stamps it on essentially every account. No loop: we defer only for names
the registrar HAS, and it never queries us for those.
Co-Authored-By: Claude Fable 5 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A recent change made this coinos.io lnurlp endpoint defer any
migrated-flagged account to the v3 names registrar and 404 when the
registrar didn't have the name. But registration stamps that flag on
essentially every account and never syncs names to the registrar, so
this broke lightning-address resolution for nearly all accounts.
The v3 registrar is the front for v3 accounts and falls back to this
endpoint, so this endpoint is the fallback target: it must serve its own
accounts directly and must not defer back (which also avoids a loop once
the registrar's fallback is live). Serve the local account; 404 only when
no local account exists.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Claude-Session: https://claude.ai/code/session_01Dpqu6wizYe1c1g4k5Gd8JK
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
The stored replaceable 13194 info event was intermittently disappearing
from the relay, leaving NWC clients (Alby etc.) with "no info event"
until the next hourly existence-check happened to republish it. Drop the
check-then-publish dance and just republish both server info events every
2 minutes — a replaceable event overwrites in place, so it's cheap and
self-heals immediately. Silence the per-publish success log since it now
fires on every interval.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Claude-Session: https://claude.ai/code/session_01Dpqu6wizYe1c1g4k5Gd8JK
|
| |
|
|
|
|
|
|
|
|
|
|
| |
A `migrated` flag on the user, settable through /user and visible on the
public user endpoint. It changes exactly one thing: lnurlp for that
username defers to the names registrar, so payments to [email protected]
arrive in the person's v3 wallet. The account itself is untouched — same
username, same history, still usable here — which is why this is a flag
and not a rename or a shutdown.
Co-Authored-By: Claude Fable 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01Wvj4aoeMZwA7VUGdJP9Swi
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Clamp a non-internal receipt's tip to the settled amount so a tip
mutated larger than the amount actually received can't produce a
negative, reversible payment record on the recipient.
Also reject an autowithdraw whose destination resolves back to the
sender: that self-send re-fires completePayment and, since the balance
nets out, never drops below threshold, looping unbounded. The route's
create() already has the equivalent self-send guard; the internal
pay() helper lacked it.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Claude-Session: https://claude.ai/code/session_01Dpqu6wizYe1c1g4k5Gd8JK
|
| |
|
|
|
|
|
|
|
| |
An address at this domain may now belong to a coinos v3 wallet rather
than a local account. When the username isn't ours, ask the names
registrar and pass its payRequest through unchanged — its callback is
absolute, so the payer talks to it directly from there.
Co-Authored-By: Claude Fable 5 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
sendKeysend debited, ran ln.keysend, and on a throw decided whether to
refund by calling listpays({payment_hash: hash}) with the CALLER's id (a
Nostr ev.id on the NWC path). keysend randomizes its own payment hash —
the CLN command takes no preimage arg — so that lookup could never match:
it returned [], `failed` was always true, and reverse() ran on EVERY
throw. A keysend that threw but actually settled was re-credited while
the sats had left the node, a repeatable hot-wallet drain. Nothing
corrected it either: unlike sendLightning, keysends are never enrolled in
`pending`, so check() never revisits them.
Recovering the real hash from the error payload doesn't work — verified
on CLN v26.06.2 that keysend errors carry no payment_hash under any key
(the error is {code, message, attempts}), so that approach never reverses
and instead eats the user's money on every ordinary routing failure.
Instead tag the payment with our own unique label and record where the
sendpays index stands first (`wait sendpays created 0` returns the
current index without blocking), so a throw can list only sendpays
created since — no full scan — and match ours by label whatever the error
looks like. CLN writes the sendpay record BEFORE sending the HTLC, so no
record at all proves nothing left the node and the debit is safe to
reverse. Verified both directions on regtest: an unroutable keysend is
refunded, and a settled one is correctly detected and left alone.
Known gap: a keysend still in flight when the RPC throws stays debited,
since keysends are not enrolled in `pending` for check() to reconcile.
Reported-by: Muhammed Shekho (@SYR-ROOT) <[email protected]>
Co-Authored-By: Claude Fable 5 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
/confirm is unauthenticated, and its first-seen branch credited a
deposit after a plain getPayment() read with no lock. Concurrent callers
(walletnotify, catchUp, bulk sweeps, or anyone posting the endpoint) all
read p == null for one deposit and each ran credit(), inflating both the
pending balance and the spendable credit:<net>:<uid> fee accumulator —
a fee waiver nobody paid for.
Verified on regtest: 10 concurrent /confirm on one 100,000 sat deposit
credited 1,000,000 pending and 4,000 fee credit. With the NX guard,
exactly 100,000 and 400. Mirrors the lock the confirm stage already had.
Reported-by: Muhammed Shekho (@SYR-ROOT) <[email protected]>
Co-Authored-By: Claude Fable 5 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
take() enforced single-use by reading authorization.claimed, checking it,
then writing it back — with yielding awaits in between and no lock. N
concurrent POST /take for one authorization all passed the gate, so each
debited the authorizer and funded the pool. Verified on regtest: 10
concurrent takes redeemed one 10,000 sat single-use authorization 5
times (50,000 debited from the authorizer, 50,000 credited to the
caller). With the SET NX claim, exactly one wins.
Funding also runs inside a try that releases the claim on failure and
only marks the record claimed once the money has landed — otherwise a
funding that throws (insufficient balance, server limit) burns the
authorization: claim key set, fund never funded, every later take
skipping the funding block.
Reported-by: Muhammed Shekho (@SYR-ROOT) <[email protected]>
Co-Authored-By: Claude Fable 5 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Tightens our implementation of the BIP-321-based NWC methods proposed
in https://github.com/nostr-wallet-connect/nwc/pull/2 — the successor
to pay_invoice/make_invoice that clients are asked to transition to
for bolt12 zaps in https://github.com/nostr-protocol/nips/pull/2421:
- reject cross-network payment instructions before paying
(UNSUPPORTED_NETWORK) instead of failing after the debit
- reject unknown req- BIP-321 params, including req-pop since we
can't open proof-of-payment callbacks
- reject a request amount that conflicts with the instruction amount
- payer_note is delivered or rejected, never dropped: bolt11+note is
rejected, internal settlements hand the note over as the payment memo
- receive mints a per-request bolt12 offer carrying the requested
description/amount instead of reusing the standing amountless offer
- get_info reports the node's actual network instead of "mainnet"
Co-Authored-By: Claude Fable 5 <[email protected]>
|
| |
|
|
|
|
|
|
| |
The old mint's keysets are redeem/melt-only — wallets must never mint or swap
into them. Force active:false in the merged /v1/keysets so mint.coinos.io
advertises a single active sat keyset (the new one) instead of two. The old mint
is panic-locked (mint/swap disabled) regardless; this fixes the client-facing
view a reviewer flagged.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Front the current mint (fresh keys, normal ops) as primary, and transparently
route pre-shutdown ecash melts to the retired panic-mode mint so wallets can melt
old ecash at mint.coinos.io without changing their URL:
- /v1/keysets is the union of both mints' keysets
- /v1/keys/<id> and melt execute route by the keyset owner
- a melt quote is taken on the primary and the invoice remembered, so an
old-keyset melt is re-quoted and executed on the recovery mint (where panic
mode gates blacklisted funds)
- /v1/checkstate is merged across both; CORS added for web wallets
The recovery mint also stands alone at mintrecovery.coinos.io.
|
| |
|
|
|
|
|
|
|
|
|
| |
The zap dispatch parsed the invoice description and, when it wasn't JSON,
suppressed the error only if the message contained "Unexpected" — V8's parse-
error wording. On Bun the message is "Unable to parse JSON string", so every
paid invoice with a non-JSON memo logged a warning. Parse-then-check instead:
a non-JSON description is simply not a zap and is skipped silently; warn only
when an actual kind-9734 zap request fails to process.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
|
| |
|
|
|
|
|
|
|
| |
A pubkey can be in `${uid}:apps` before its `app:<pubkey>` record is written
(NWC-connection creation race) or can outlive its record. Destructuring `notify`
off the null lookup threw, and since the catch wraps the whole loop it aborted
notifications for ALL of the user's apps. Skip null/absent apps and continue.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
| |
- meltguard: out-of-band CLNRest spend limiter (per-melt cap + cumulative
hard-freeze) that sits between the panic-mode mint and the node
- mintdemux: fronts the mint hostname, routing between the retired and current
mint instances after the hostname swap
- gitignore *.bak / *.bak-* so local backups (some hold secrets) never commit
Co-Authored-By: Claude Opus 4.8 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- invoices: reject caller-supplied bolt11/bolt12; the server always mints its
own. Closes the invoice-registration path behind the shared-node credit and
payment-hijack vectors.
- lightning: skip mint-owned invoices (lbl* label) in the paid-invoice listener.
- payments: lightning free-tier fee credit, withheld for flagged high-throughput
(ln:nofree) accounts which pay an elevated rate; the failed-send reversal
restores credit at the same rate. freezeCheck made crash-resilient. Lightning
and bolt12 sends stay pending until the preimage settles.
- payments (route): validate payment-request / invoice input types.
- logging: collapse repeated xpay failure lines into one.
Co-Authored-By: Claude Opus 4.8 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
| |
createproof's default proof omits invreq_payer_note, and passing include
replaces the default disclosure set — so for zap payments request both
invreq_payer_note and invoice_amount, which NIP-177 verification requires.
Verified end to end on v26.06.2: NWC pay over nip44 returns a proof that
decodes valid with note + amount disclosed and preimage matching the
payment hash.
Co-Authored-By: Claude Fable 5 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- sendLightning fetches an invoice from lno1 offers (payer_note passthrough
for NIP-177 zap intents) and pays it through the existing bolt12 path
- NWC: new nwc#2 pay/receive methods (BIP-321 URIs, payer proofs stubbed
until CLN 26.06 createproof), nip44_v2 alongside nip04 with per-request
scheme detection, dual 23196/23197 notifications, 13194 refreshed when
its content or encryption tag goes stale
- standing per-user bolt12 offer via GET /offer; incoming bolt12 payments
record invreq_payer_note as memo
- parse/fetchinvoice routes handle offers; zap display parses kind 9736
receipts (amount tag + embedded intent) and drops a call to a sync()
helper that never existed
Co-Authored-By: Claude Fable 5 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
POST /preimages now accepts fiat + optional currency (defaults to the
merchant's) as an alternative to a sat price. The fiat amount is stored
per entry and converted to sats at the moment a usePreimage invoice is
created, so the price tracks the exchange rate instead of freezing at
upload time. The invoice also takes on the entry's currency and rate so
receipts display the fiat price exactly as the merchant set it. If no
rate is available for the entry's currency, the preimage is pushed back
onto the queue and creation fails rather than issuing an any-amount
invoice.
Co-Authored-By: Claude Fable 5 <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Invoice creation is unauthenticated, so buyers could name their own
price for a queued preimage (create the usePreimage invoice with
amount: 1). POST /preimages now takes an optional price (sats) that is
stored per entry and overrides the requested amount when the preimage
is consumed; omitting it keeps the old any-amount behavior. Also gate
the stored-preimage reveal in credit() on the invoice being fully paid,
since internal sends can settle partial amounts. Bare-hex entries
queued before this change still parse (treated as priceless).
Co-Authored-By: Claude Fable 5 <[email protected]>
|
| |\
| |
| |
| |
| | |
# Conflicts:
# lib/payments.ts
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Merchants POST /preimages (auth) to pre-load 32-byte hex preimages into a
FIFO queue; creating an invoice with usePreimage: true consumes the next
one so the invoice's payment hash is sha256(preimage). Paying the invoice
reveals the merchant-chosen secret to the buyer — via their own wallet on
external settlements, and via GET /invoice/:id once settled — enabling
backend-less digital download stores (encrypted files on static hosting,
preimage + public string = decryption key).
- GET /preimages lists the queue with payment hashes so the merchant can
map files to future invoices; POST /preimages/delete clears it
- empty queue fails invoice creation rather than issuing a random
preimage the buyer would pay for but couldn't decrypt with
- preimages cln rejects as already used are discarded and the next tried
- consumed preimages persist at preimage:${payment_hash} (TTL = invoice
expiry) and credit() prefers them as the settlement ref, so internal
coinos-to-coinos payments reveal the real preimage instead of the
sender's uid
Co-Authored-By: Claude Fable 5 <[email protected]>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
apiv2.nobitex.ir became unreachable from our egress, and it was fetched
inline on every binance tick, so it failed and logged on every tick.
Move the Iranian market-rate fetch onto its own 60s timer with a request
timeout, cache the last-good value (falling back to the fixer rate after
an hour), and throttle failure logging. Switch the source to Wallex
(api.wallex.ir BTCTMN), which is reachable.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
On-chain deposits below 300 sats are dropped by the dust filter in the
confirm/credit path, so a specific-amount bitcoin/liquid invoice under
that threshold can never be credited. Reject creating one at generation
time and point the user at lightning. Any-amount (0) addresses are
unaffected.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| | |
| |
| |
| |
| |
| |
| |
| | |
Add a per-IP rate limit (30/10s) to POST /invoice and POST /invoice/:id,
matching the existing GET /invoice/:id limit, to throttle automated
invoice-creation floods.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| | |
| |
| |
| |
| |
| |
| |
| | |
The preHandler request logger wrote the full request body, which included
plaintext passwords on auth requests. Redact password/confirm fields
before logging.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
credit() now credits the tip the sender was actually debited (passed in
by the internal-transfer callers) instead of independently re-reading the
invoice. This closes a time-of-check/time-of-use gap where an invoice's
tip could be changed between the sender debit and the receiver credit,
letting a payer be credited a tip they were never charged.
External receipts (lightning/ecash) keep reading the invoice tip, since
there is no paired coinos debit to diverge from.
Also harden POST /invoice/:id: validate the tip is a sane non-negative
integer and reject tip changes on an already-settled invoice.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
| |
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
|
| |
The nobitex endpoint host changed (old host now NXDOMAIN) and the fetch
was swallowed by an empty catch, silently blanking Iranian rates. Point
at the current host, and since the API quotes in Rial despite the IRT
pair name, set IRR from lastTradePrice and IRT = Rial/10. Log fetch
failures instead of ignoring them.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
publishInfo sent the kind-13194 info event over the long-lived NWC socket
fire-and-forget. The nostr lib Relay.send is async (await wait_connected; ws.send)
but was called without await, and success was logged immediately. A send that
rejected (e.g. ws in CLOSING state during a strfry reconnect) was silently
dropped while the log claimed success, leaving NWC clients with "no info event"
until a manual republish.
Publish over a fresh short-lived socket and wait for the relay OK, retrying up to
3x with honest logging (warn on unconfirmed, err on give-up). ensureInfos
receive-based detection is unchanged. Verified on prod: deleted both info events,
restarted, both republished within seconds.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
| |
Emit a greppable EVICTED_AUTH line (username + cf-connecting-ip) when an
evicted account's JWT is used, so the real-time auto-banner can block the
source IP at the edge immediately rather than waiting for the hourly scan.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
|
| |
Add an `evicted` set check in the JWT strategy: an evicted account fails
authentication on every request and every endpoint, regardless of source
IP. The existing blacklist only blocks sends, so a compromised/abusive
JWT could still authenticate and read/probe; eviction kills it outright.
Matches on uid or username so a rename can't evade it.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
| |
The admin-only password-reset endpoint returned "disabled" to non-admin
callers, which confirmed both that the endpoint exists and that it is
admin-gated. Return a generic 404 identical to an unregistered route
instead; still log the rejected attempt server-side for monitoring.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
BOLT11 is case-insensitive, but an uppercase-pasted invoice was stored as
the payment hash and the pending-set entry verbatim. check()'s
startsWith("ln") test is case-sensitive, so it treated such pending
entries as non-lightning and evicted them without reversing — a failed
send would never be auto-refunded.
- sendLightning: lowercase the invoice up front so the stored hash, the
pending entry, and xpay all use the same form.
- check(): lowercase before the startsWith("ln") test (defensive; also
rescues any already-stored uppercase entries).
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
check() skipped pending payments younger than 10s, but xpay retries for
30s (retry_for: 30). During that window listpays can transiently show all
parts failed before the winning part lands, so check() could reverse a
payment sendLightning was still driving; the part then completed and
finalize() ran against an already-removed record.
- check(): raise the skip threshold 10s -> 60s so sendLightning finishes
finalize()/reverse() and removes the payment from `pending` before
check() ever touches it.
- finalize(): if the record is gone, raise a clear error instead of a
cryptic null-deref.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
|
| |
The source fix (LPOS guard in generate()) stops NEW duplicate entries but
doesn't remove the ones already accumulated. This one-time script scrubs
them: scans every <id>:invoices list and removes duplicate ids, keeping
each id's original creation-order position. Dry-run by default; pass
--apply to write. Idempotent.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
| |
generate() unconditionally lPush'd the invoice id to `${aid}:invoices`,
but it's also called from the update() handler (tip/webhook changes, and
repeated bolt12 offer creation) re-using the existing id — so each update
re-appended the same id and the user's :invoices accumulated duplicate
entries, rendering the same offer/invoice multiple times. Only push when
the id isn't already in the list (LPOS guard).
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The /lnurl/proxy, /decode and /encode handlers fetched an arbitrary
user-supplied URL via got() and returned the body — an unauthenticated
SSRF reaching internal services. Route those fetches through a new
safeGot() that rejects hosts resolving to loopback/private/link-local/
CGNAT/metadata ranges, re-validates on each redirect hop, restricts the
scheme to http(s), and caps timeout + response size. Also validate the
NIP-57 zap request (kind 9734 + signature) before storing it, instead of
trusting the client-supplied event.
These endpoints can't be removed — they back lightning-address/LNURL
parsing and resolution in the wallet UI.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
| |
After broadcasting an on-chain send via bitcoind, also POST the raw tx to
mempool.space and blockstream (fire-and-forget, non-blocking) so those
explorers see it immediately instead of waiting ~8s for p2p propagation.
Skipped on regtest/testnet.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| | |
|
| |
|
|
|
|
|
|
|
| |
audit-wiped-history.ts (read-only) finds accounts whose :payments index
was wiped by the changeid bug — balance>0 with empty history in both db
and arc. rebalance-economics.ts models whether the lightning->liquid
bridge throughput is fee-positive against on-chain rebalance cost.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
|
| |
POST /delete (auth) wipes every key tied to the calling user, mirroring
scripts/delall-legacy.ts: user records, payments/invoices, sub-accounts,
NWC apps, nostr follow graph + nip5. Gated by typing the username and
refuses while balance+pending > 10k sats so funds can't be destroyed by
accident.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
|
| |
getPayment() pre-check in the lightning listener/replay isn't atomic, so
concurrent re-credits of the same settlement each ran incrBy. Claim each
lightning/bolt12 settlement with an atomic SET NX on credited:<ref>
(ref = preimage) and bail if already claimed. Also disable the /replay
route that triggered it.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
| |
|
|
|
|
|
|
|
| |
The blacklist (send freeze) only matched the current username, so a
compromised account could shake the freeze by renaming itself. Match on
either the lowercased username OR the immutable uid; add the uid to the
`blacklist` set to freeze durably.
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
|
