feat(identity): seed-derivation verifier + KAT; rename "Your DID"→"Node DID"

- scripts/verify-seed-derivation.py: stdlib-only tool to cryptographically prove
  a node's on-disk keys (node_key→DID, nostr_secret→npub, fips_key) are derived
  from its onboarding seed exactly as seed.rs documents (BIP-39 → PBKDF2-HMAC-
  SHA512 → HKDF-SHA256 with per-key domain separation).
- seed.rs: known-answer regression test cross-checking Rust node_key + nostr
  bytes against the Python verifier (locks the derivation).
- en.json: "Your DID" → "Node DID".

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
archipelago 2026-06-16 10:17:29 -04:00
parent 67609eea91
commit 3a9d1db763
3 changed files with 148 additions and 1 deletions

View File

@ -543,4 +543,22 @@ mod tests {
}
}
}
#[test]
fn test_node_key_known_answer_vs_python_verifier() {
// Cross-checks scripts/verify-seed-derivation.py: same mnemonic must
// produce the same node_key bytes in Rust and in the Python verifier.
let (_, seed) = MasterSeed::from_mnemonic_words(TEST_MNEMONIC).unwrap();
let key = derive_node_ed25519(&seed).unwrap();
assert_eq!(
hex::encode(key.to_bytes()),
"3b4f4a1450450260ae360adb9c33ea5eb86356fa14454ca0067dd4b51ea8be87"
);
let nostr = derive_node_nostr_key(&seed).unwrap();
assert_eq!(
hex::encode(nostr.secret_key().to_secret_bytes()),
"3a94fb32efab2a5025401d53fd7d82b41323a5c06ad14ce528ebe3a813d88831"
);
}
}

View File

@ -179,7 +179,7 @@
"aiDataAccess": "AI Data Access",
"serverName": "Hostname",
"sessionStatus": "Session Status",
"yourDid": "Your DID",
"yourDid": "Node DID",
"onionAddress": "Node .onion Address",
"loggedIn": "Currently logged in",
"didHelper": "Decentralized identifier for passwordless auth",

View File

@ -0,0 +1,129 @@
#!/usr/bin/env python3
"""
Cryptographically verify that a node's on-disk keys are deterministically
derived from its onboarding seed, exactly as documented in core/archipelago/
src/seed.rs:
BIP-39 mnemonic (24 words)
-> PBKDF2-HMAC-SHA512(2048, salt="mnemonic") = 64-byte seed
-> HKDF-SHA256(salt=None, IKM=seed, info=<domain>) = each 32-byte key
"archipelago/node/ed25519/v1" -> node_key (=> Node DID)
"archipelago/nostr-node/secp256k1/v1" -> nostr_secret (=> npub)
"archipelago/fips/secp256k1/v1" -> fips_key (FIPS transport)
It compares each freshly-derived key against the bytes actually on disk under
/var/lib/archipelago/identity/. A MATCH proves the on-disk key was derived from
the seed (and nothing else). Also prints the resulting did:key for cross-check
against Settings -> Node DID.
Usage (run on the node):
sudo python3 verify-seed-derivation.py
# paste the 24-word mnemonic when prompted (input is hidden, never logged)
Pure standard library no third-party crypto packages required.
"""
import sys, os, hmac, hashlib, getpass, unicodedata
IDENT = "/var/lib/archipelago/identity"
DOMAINS = {
"node_key (=> Node DID)": (b"archipelago/node/ed25519/v1", f"{IDENT}/node_key", "raw"),
"nostr_secret (=> node npub)": (b"archipelago/nostr-node/secp256k1/v1", f"{IDENT}/nostr_secret", "nsec"),
"fips_key (FIPS transport)": (b"archipelago/fips/secp256k1/v1", f"{IDENT}/fips_key", "nsec"),
}
def hkdf_sha256(ikm: bytes, info: bytes, length: int = 32) -> bytes:
"""RFC 5869 HKDF-SHA256 with salt=None (== HashLen zero bytes)."""
salt = b"\x00" * hashlib.sha256().digest_size
prk = hmac.new(salt, ikm, hashlib.sha256).digest()
okm, t, i = b"", b"", 1
while len(okm) < length:
t = hmac.new(prk, t + info + bytes([i]), hashlib.sha256).digest()
okm += t
i += 1
return okm[:length]
# --- minimal bech32 decode (BIP-173) to recover the 32-byte secret from nsec ---
_B32 = "qpzry9x8gf2tvdw0s3jn54khce6mua7l"
def _bech32_decode_data(s: str) -> bytes:
s = s.strip().lower()
pos = s.rfind("1")
data = [_B32.index(c) for c in s[pos + 1:]]
data = data[:-6] # drop 6-char checksum
# convert 5-bit groups -> 8-bit bytes
acc = bits = 0
out = bytearray()
for v in data:
acc = (acc << 5) | v
bits += 5
if bits >= 8:
bits -= 8
out.append((acc >> bits) & 0xFF)
return bytes(out)
# --- minimal base58btc + multicodec to render did:key from the ed25519 pubkey ---
_B58 = "123456789ABCDEFGHJKLMNPQRSTUVWXYZabcdefghijkmnopqrstuvwxyz"
def _b58(b: bytes) -> str:
n = int.from_bytes(b, "big")
s = ""
while n:
n, r = divmod(n, 58)
s = _B58[r] + s
return "1" * (len(b) - len(b.lstrip(b"\x00"))) + s
def did_key_from_ed25519_pub(pub: bytes) -> str:
return "did:key:z" + _b58(b"\xed\x01" + pub) # 0xed01 = ed25519-pub multicodec
def main() -> int:
if not os.path.isdir(IDENT):
print(f"!! {IDENT} not found — run this on a node.")
return 2
mnemonic = getpass.getpass("Paste the node's 24-word mnemonic (hidden): ").strip()
words = mnemonic.split()
if len(words) != 24:
print(f"!! expected 24 words, got {len(words)}")
return 2
# BIP-39: seed = PBKDF2-HMAC-SHA512(NFKD(mnemonic), "mnemonic"+passphrase, 2048, 64)
norm = unicodedata.normalize("NFKD", " ".join(words)).encode("utf-8")
seed = hashlib.pbkdf2_hmac("sha512", norm, b"mnemonic", 2048, 64)
all_ok = True
for name, (info, path, fmt) in DOMAINS.items():
derived = hkdf_sha256(seed, info, 32)
try:
raw = open(path, "rb").read()
disk = raw if fmt == "raw" else _bech32_decode_data(raw.decode().strip())
disk = disk[:32]
except Exception as e:
print(f"[{name}] could not read {path}: {e}")
all_ok = False
continue
ok = disk == derived
all_ok &= ok
print(f"[{'MATCH ✅' if ok else 'MISMATCH ❌'}] {name}")
print(f" derived(seed): {derived.hex()}")
print(f" on-disk : {disk.hex()}")
# Render the Node DID from node_key.pub for a visual cross-check vs the UI.
try:
pub = open(f"{IDENT}/node_key.pub", "rb").read()[:32]
print(f"\nNode DID (from node_key.pub): {did_key_from_ed25519_pub(pub)}")
print(" ^ should equal Settings -> Node DID")
except Exception:
pass
print("\n==> ALL KEYS SEED-DERIVED ✅" if all_ok else "\n==> SOME KEYS DID NOT MATCH ❌")
return 0 if all_ok else 1
if __name__ == "__main__":
sys.exit(main())