Implement session call digest for v3 sessions#362
Merged
Conversation
ScreamingHawk
requested changes
Jul 13, 2026
ScreamingHawk
left a comment
ScreamingHawk
left a comment
Contributor
There was a problem hiding this comment.
The old hashCallWithReplayProtection was replaced with hashPayloadCallIdx in the contract as per the PR text. I suggest renaming this function to match the new name too.
Otherwise LGTM!
taylanpince
approved these changes
Jul 13, 2026
ScreamingHawk
added a commit
that referenced
this pull request
Jul 13, 2026
#362 was merged with the digest function renamed to HashPayloadCallIdx; update the parent-wallets test to the new name so the package builds. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
ScreamingHawk
added a commit
that referenced
this pull request
Jul 19, 2026
* fix(v3): sync session encoding with wallet-contracts-v3 The explicit-session and implicit-attestation encodings had drifted from the deployed v3 contracts, so any config image hash / session signature go-sequence produced was rejected on-chain. Re-sync three changes the contracts made (and mirror the sequence.js primitives JSON): - SessionPermissions: add chainId (32 bytes, after signer) and encode deadline as uint64 (8 bytes) instead of 32. Matches SessionSig recoverConfiguration and permission.ts. - Attestation authData: append issuedAt (uint64, 8 bytes) so the attestation hash matches LibAttestation.toHash. Serialize issuedAt as a decimal string in JSON to match attestation.ts. - Wire issuedAt through the cmd/sequence CLI input. Add regression tests with byte vectors verified against the contracts via forge (recoverConfiguration / LibAttestation.toHash). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(v3): drop session wallet from call digest parent wallets HashCallWithReplayProtection hashed the payload's parentWallets verbatim. The session-holding wallet is the last parent wallet in the signing context, but on chain it is the verifying contract (msg.sender), not a parent entry, so it must be dropped before hashing to match SessionSig.hashPayloadCallIdx. Wallet-bound digest comes from the partial payload replay audit fix (wallet-contracts-v3 9c19609, Code4rena S-93/S-169, #89). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(v3): validate session deadline fits in uint64 EncodeSessionPermissions encodes deadline into an 8-byte buffer via big.Int.FillBytes, which panics on a negative value or one that does not fit in uint64 (e.g. a max-value "no expiry" deadline). Validate the range and return an error instead of crashing. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(v3): validate numeric issuedAt in attestation JSON AttestationFromJson accepted a JSON-number issuedAt via an unchecked uint64 cast, so a negative, fractional or out-of-range value was silently coerced (e.g. -1 -> MaxUint64) and the library would hash/sign a different attestation than the JSON represented. Reject negatives, fractions and values outside uint64, matching the string path. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * test(v3): align parent-wallets test with HashPayloadCallIdx rename #362 was merged with the digest function renamed to HashPayloadCallIdx; update the parent-wallets test to the new name so the package builds. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> * fix(v3): decode session/attestation numbers safely, validate ranges Numeric JSON attributes were decoded and encoded unsafely: an unquoted number above 2^53 was silently rounded by encoding/json before use, and a negative or over-wide chainId/valueLimit was encoded as its absolute value or spilled past the fixed 32-byte field. Add bigIntFromJSON, a single decoder shared by chainId, valueLimit, deadline and attestation issuedAt: it parses strings and json.Number exactly and rejects float64 that is non-integer or at/above 2^53. EncodeSessionPermissions now rejects chainId and valueLimit that are negative or wider than uint256. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This implements the digest that a session key has to sign for each call of a v3 payload, needed for waas smart-sessions (7193). The old
hashCallWithReplayProtectionwas a leftover stub, unexported and with no callers, so I replaced it with an exportedHashCallWithReplayProtectionthat matchesSessionSig.hashPayloadCallIdxin wallet-contracts-v3:keccak256(Payload.hashFor(payload, wallet) ++ uint256(callIdx)).Note that the digest scheme the stub hinted at (
chainId ++ space ++ nonce ++ callHash) is no longer what the contracts verify, it was removed in the audit fix for partial payload replay (0xsequence/wallet-contracts-v3#89), so implementing it as described would produce signatures that never pass on-chain.The test vectors in
session_digest_test.gowere generated fromSessionSig.hashPayloadCallIdxitself, with a forge test constructing identical payloads, so the test asserts parity with the on-chain verification. I also checked that a signature produced over this digest recovers to the right signer through plainecrecover, which is the exact path the contract uses (no EIP-191 prefix).