Skip to content

feat: add HelloChrome_155 and make it HelloChrome_Auto - #4

Merged
Nemu-x merged 1 commit into
devfrom
feat/chrome155
Nov 10, 2026
Merged

Nemu-x merged 1 commit into
devfrom
feat/chrome155

Conversation

@Nemu-x

@Nemu-x Nemu-x commented Nov 10, 2026

Copy link
Copy Markdown
Collaborator

Ports the Chrome 155 profile from refraction-networking/utls 6ebdceb ("Add Chrome 155 fingerprint verified against a fresh browser capture", verified there against Chrome 155.0.8059.40 on macOS). Needed for Citrus to make client-fingerprint: chrome mean Chrome 155, as Xray-core 26.11.9 did.

What changes on the wire (vs Chrome 133)

  • signature_algorithms: a GREASE value first, then ML-DSA-44/65/87 (0x0904/0x0905/0x0906), then the old eight schemes.
  • New trust_anchors extension (0xca34) with the 28 Chrome Root Store IDs from the capture.
  • Everything else (ciphers, groups, X25519MLKEM768 + X25519 key shares, new ALPS codepoint, GREASE ECH, extension shuffling) is the same as 133.

Changes

  • HelloChrome_155 + spec in u_parrots.go; HelloChrome_Auto now points to 155 (upstream did the same). HelloChrome_133 is untouched.
  • TrustAnchorsExtension (u_trust_anchors.go), wired into ExtensionFromID, so the Fingerprinter and JSON give a typed extension.
  • GREASE inside signature_algorithms now uses its own BoringSSL seed slot (slot 7). The internal seed array grows to 8 slots. The exported GetBoringGREASEValue([5]uint16, int) keeps its signature.
  • MLDSA44/65/87 SignatureScheme constants (+ stringer output) and dicttls names. Difference from upstream: upstream had real ML-DSA support from its go 1.27 sync. This fork does not, so the codepoints are only advertised. If a server ever picked ML-DSA, the handshake would fail with "unsupported signature algorithm". REALITY/JLS servers and normal web servers do not pick it.

Tests

Ported from upstream: the captured ClientHello fixture (testdata/chrome155_clienthello.hex, 1955-byte record) compared with the generated hello (normalizing only GREASE, extension order, key material and ECH randomness), the Fingerprinter round-trip, the signature-GREASE seed slot, Chrome 133 unchanged, the trust-anchors encode/decode/JSON/limits checks, ML-DSA JSON names, and the GREASE public API.
Not ported: TestChrome155MLDSAHandshake (needs crypto/mldsa) and TestChrome155PresetOwnership (needs upstream's ClientHelloSpec.clone, which this fork doesn't have; the fresh trust-anchor slices are still returned per spec).

go test ./... and go vet pass with go1.26.10 (windows/amd64).

Cross-check: the fixture is byte-identical to the one in Rerowros/veles (component/tls/testdata/chrome155_clienthello.hex), and that project's hand-written spec has the same contents. It is the same capture, though, not an independent one.

Port the Chrome 155 profile from refraction-networking/utls 6ebdceb
("Add Chrome 155 fingerprint verified against a fresh browser capture"),
verified there against Chrome 155.0.8059.40.

Compared with Chrome 133 the ClientHello gains a GREASE value and the three
ML-DSA schemes (0x0904/0x0905/0x0906) in signature_algorithms and the
trust_anchors extension (0xca34) with 28 Chrome Root Store IDs.

- HelloChrome_155 and its spec; HelloChrome_Auto now points to it, as
  upstream does. HelloChrome_133 is unchanged.
- TrustAnchorsExtension (typed, JSON, Fingerprinter via ExtensionFromID).
- GREASE in signature_algorithms drawn from its own BoringSSL seed slot;
  the internal seed grows to 8 slots while GetBoringGREASEValue keeps its
  [5]uint16 signature.
- MLDSA44/65/87 SignatureScheme constants and dicttls names. Unlike
  upstream (go 1.27 sync) this fork does not implement ML-DSA, so the
  codepoints are only advertised.
- Upstream tests and the captured ClientHello fixture. Not ported: the
  ML-DSA handshake test (needs crypto/mldsa) and the preset ownership test
  (needs upstream's ClientHelloSpec.clone).
@Nemu-x
Nemu-x merged commit 870e241 into dev Nov 10, 2026
3 of 4 checks passed
@Nemu-x
Nemu-x deleted the feat/chrome155 branch November 10, 2026 22:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant