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.
45 lines
1.9 KiB
Docker
45 lines
1.9 KiB
Docker
# hushd. Built amd64 IN-CLUSTER by Kaniko — never `docker build` locally, which
|
|
# on an Apple Silicon laptop produces an arm64 image the cluster cannot run.
|
|
#
|
|
# The build is HERMETIC: `-mod=vendor` with `GOPROXY=off` and `GOFLAGS=-mod=vendor`
|
|
# means no module download and no network. That is not an optimisation — it is
|
|
# required, because github.com/orchard9/go-chassis is a private module and the
|
|
# build container holds no git credential. `make vendor` is what moves versions.
|
|
FROM golang:1.26-alpine AS build
|
|
WORKDIR /src
|
|
|
|
# Vendored, so this is the whole dependency graph: no go.sum verification step,
|
|
# no proxy, no cache warm-up.
|
|
COPY go.mod go.sum ./
|
|
COPY vendor/ ./vendor/
|
|
COPY cmd/ ./cmd/
|
|
COPY internal/ ./internal/
|
|
|
|
# Fail loudly if anything reaches for the network, rather than silently falling
|
|
# back to a proxy that will not be there in CI.
|
|
ENV GOFLAGS=-mod=vendor GOPROXY=off CGO_ENABLED=0
|
|
|
|
# Tests run in the CI step, not here: a Kaniko layer that runs tests caches
|
|
# their result and stops re-running them. Keep the image build to building.
|
|
RUN go build -trimpath -ldflags="-s -w" -o /out/hushd ./cmd/hushd
|
|
|
|
# distroless static + nonroot: no shell, no package manager, no libc surface.
|
|
# hushd needs only TCP and the CA bundle distroless already carries.
|
|
FROM gcr.io/distroless/static-debian12:nonroot AS runtime
|
|
WORKDIR /
|
|
|
|
COPY --from=build /out/hushd /hushd
|
|
|
|
# Numeric, not "nonroot": with a non-numeric USER the kubelet cannot verify
|
|
# runAsNonRoot and refuses to start the pod.
|
|
USER 65532:65532
|
|
|
|
# HTML templates are embedded in the binary (internal/web, go:embed), so there
|
|
# is no asset directory to mount, drift, or go missing at runtime.
|
|
EXPOSE 18500
|
|
|
|
# Exec form so SIGTERM reaches the process directly. The chassis's two-phase
|
|
# drain (readiness 503, wait, then shutdown) depends on receiving it, and the
|
|
# Deployment's terminationGracePeriodSeconds is sized to outlast that drain.
|
|
ENTRYPOINT ["/hushd"]
|