4d9a26498e
Paste a secret, get a link, send it. The first person to open it and press
Reveal sees the secret; the link dies at that moment. The recipient needs a
browser and nothing else — no account, no client, no installed tooling.
The server cannot read what it stores. AES-256-GCM happens in the browser and
the key lives in the URL fragment, which browsers never transmit, so hushd
holds ciphertext and no key material. That is a property of where the key sits
rather than a promise about our conduct, which is why there is deliberately no
endpoint accepting a plaintext secret and no server-side-encryption fallback:
two guarantees behind one URL would be worse than one honest guarantee.
Three decisions carry the design:
* GET /s/{id} touches NO storage, not even to check existence. Slack, Teams,
WhatsApp, iMessage and Outlook Safe Links all fetch a URL before a human
sees it, so destroying on GET would destroy most secrets in transit and the
recipient's "already used" would be indistinguishable from interception.
Only POST /reveal consumes. Bot user-agent detection is an arms race;
removing the side effect from GET is not. Pinned by
TestGettingTheRevealPageNeverConsumesTheSecret.
* Destruction is one Redis GETDEL, which is atomic. GET-then-DEL has a window
where two simultaneous readers both win, and for a one-time secret that
window is the product. The store contract demands atomicity and the same
concurrency test runs against both implementations.
* Missing, already-revealed, expired and evicted are ONE indistinguishable
410. Separating them would confirm to a prober that a given link was real.
The secret id IS the capability, so secret.ID is a struct whose every
accidental path — %v, %s, String(), slog, json.Marshal — emits a redacted
handle or refuses, and the raw value needs an explicit Value(). The first
version tried to prevent leaks by implementing no String() at all; its own test
caught that Go's fmt prints unexported fields anyway, so forbidding the method
had removed the control rather than the leak.
Operationally: structured JSON on stdout in the fleet's wire format, which
Vector already collects with no annotation; six hush_* metrics on the chassis
registry with no id, IP or path in any label; five alert rules wired into
vmalert. The public Ingress enumerates /, /s/ and /api/ so /metrics, /healthz
and /readyz share the port but are unreachable from the internet — no
basic-auth middleware to maintain and get wrong.
Dependencies are vendored because go-chassis is private: the Woodpecker test
step and the in-cluster Kaniko build both run -mod=vendor with GOPROXY=off and
hold no git credential.
cmd/hush-mcp is a stdio MCP server doing the same client-side crypto locally,
so using hush from an agent preserves the same guarantee as using it from a
browser.
51 lines
2.1 KiB
Go
51 lines
2.1 KiB
Go
// Package store persists ciphertext for at most a TTL and destroys it on the
|
|
// first successful read.
|
|
package store
|
|
|
|
import (
|
|
"context"
|
|
"errors"
|
|
"time"
|
|
|
|
"github.com/orchard9/hush/internal/secret"
|
|
)
|
|
|
|
// ErrGone is returned by Take for every reason a secret is not available: it
|
|
// never existed, it was already revealed, it expired, or Redis evicted it early
|
|
// under memory pressure.
|
|
//
|
|
// The causes are deliberately NOT distinguished. Not because they are hard to
|
|
// tell apart, but because telling them apart is the leak: a caller who can
|
|
// separate "never existed" from "already revealed" can confirm that a given
|
|
// link was real, which is information about someone else's secret. The API maps
|
|
// this one error to one 410 for all four cases.
|
|
var ErrGone = errors.New("secret is gone")
|
|
|
|
// ErrIDCollision means Put found the key already occupied. At 256 bits of id
|
|
// entropy this cannot happen by chance, so it means a bug or a repeated id, and
|
|
// it is surfaced rather than silently overwriting a live secret.
|
|
var ErrIDCollision = errors.New("secret id already exists")
|
|
|
|
// Store is the whole persistence contract. Two methods, both destructive-safe:
|
|
// there is no Get, no List, and no Exists — a store that could answer "does
|
|
// this id exist" without consuming it would let the reveal page leak existence,
|
|
// and a store that could list would make a compromise catastrophic instead of
|
|
// merely bad.
|
|
type Store interface {
|
|
// Put writes ciphertext under id, expiring after ttl. It must fail rather
|
|
// than overwrite an existing key.
|
|
Put(ctx context.Context, id secret.ID, ciphertext string, ttl time.Duration) error
|
|
|
|
// Take returns the ciphertext and destroys it ATOMICALLY. Two concurrent
|
|
// callers must not both receive a value; exactly one gets it and the other
|
|
// gets ErrGone. This atomicity is the one-time property — an implementation
|
|
// that reads then deletes has a window in which both callers succeed.
|
|
Take(ctx context.Context, id secret.ID) (string, error)
|
|
|
|
// Ping reports whether the backing store is usable, for readiness.
|
|
Ping(ctx context.Context) error
|
|
|
|
// Close releases resources.
|
|
Close() error
|
|
}
|