Changelog

• PhantasmaPhoenix.RPC 0.6.0 and PhantasmaPhoenix.NFT 0.6.0 are now available on NuGet. Developers using either package should update to the current versions.
• JSON-RPC request and response ids now support the protocol-valid forms: integer ids, string ids, null ids, and omitted ids.
• RPC responses now serialize the exact id value supplied by the caller instead of relying on an implicit default response id.
• Invalid JSON-RPC id shapes such as objects, arrays, and floating-point values are rejected during deserialization.
• PhantasmaPhoenix.NFT 0.6.0 now depends on PhantasmaPhoenix.RPC 0.6.0, so NFT integrations use the same corrected RPC id handling.

PhantasmaPhoenix.RPC | PhantasmaPhoenix.NFT

dotnet package update PhantasmaPhoenix.RPC

dotnet package update PhantasmaPhoenix.NFT

• LMDB read scopes are now kept short around the local database read and are no longer held across awaited network calls.
• The stability fix covers account NFT reads, single NFT reads, NFT batch reads, token NFT pagination, and transaction lookup with recent-transaction fallback.
• Extended token responses now temporarily tolerate the historical GFLP tokenSchemas row with trailing bytes, so clients can still receive schema data while the historical row remains in storage.
• No client API change is required for this update.

• pha-decode tx --hex now accepts three exact input shapes in one path: Carbon signed transactions, full VM transaction hex, and raw VM script hex.
• Carbon Phantasma wrappers are now reconstructed into output.vm, so wrapped execution can be inspected without manually unpacking the envelope first.
• Phantasma_Raw payloads now decode the inner VM transaction exactly instead of being treated as opaque bytes.
• tx --hash still prefers carbonTxData, but now falls back to top-level RPC script data when full VM reconstruction is not possible.
• Input detection was tightened so raw VM hex is no longer misclassified as Carbon data just because the prefix happens to look valid.

pha-decode

npm i -g pha-decode

• Nexus.AttachTokenContract now uses the same registration path as fresh token-backed contract creation.
• Attached token contracts now run their constructor during registration, so storage-backed fields such as _owner are initialized before owner checks, trigger execution, or getter calls read them.
• This fixes attached contracts that immediately read constructor-initialized state after attach.

https://deploy-devnet.phantasma.info/

• The update restores Gen2 VM parity for BigInteger, VMObject, ScriptContext, storage/state interops, runtime getters, and legacy address resolution.
• Contract lifecycle code is now installed on mainnet for custom and token-backed contracts, including deploy, upgrade, onUpgrade, and contract-owned NFT flows. Contract deployment and execution will remain unavailable until the activation step is completed.
• Token lifecycle support was restored in the installed code for mint, burn, send, receive, infuse, and series operations. Runtime.MintTokens and Runtime.GetTokenOwner are also included.
• NFT handling now uses deterministic Phantasma-compatible IDs and preserves imported NFT and series identities across mint, burn, transfer, and read paths.
• Chain data repair resolutions were included for imported-state restoration: contract imports restore globals, maps/lists, token-contract indexes, missing script/ABI, and persisted imported series contract bytes.
• Additional repair resolutions cover missing series and mint counts, imported token-contract fungible balances, imported market NFTs, and imported NFT infusions.
• Market and token data compatibility was repaired for imported series royalties, governance-controlled early market cancellation, Token.GetBalances, and system-address transfer semantics.
• Public RPC reads for accounts, balances, owned tokens, NFTs, and multi-result InvokeRawScript responses were corrected as part of the same mainnet rollout.
• Fresh validator and audit-node sync were improved so nodes can recover and continue replay from partially downloaded state.