fix(wallet): mint slash-free tiny.place identity keys#4779
Conversation
The tiny.place backend serves the Ed25519 identity public key as the
Signal bundle `identityKey` (STANDARD base64) and keys its
`/keys/{identityKey}/...` routes on it verbatim. A `/` in the base64
becomes `%2F` in the request path and the route 404s — so an identity
whose public key contains `/` can neither publish its Signal pre-keys
nor receive DMs. Our tiny.place identity is deterministic from the
Solana wallet key, so ~half of new wallets mint an unusable identity.
Derive a slash-free identity seed at the point we hand the seed to
LocalSigner:
- counter 0 == the raw wallet seed, so every already-clean identity is
byte-identical (the common case is unchanged, no migration),
- only when the raw key's base64 contains `/` do we deterministically
bump a domain-separated SHA-256 counter until the public key is
slash-free. Same wallet always maps to the same lowest-counter
slash-free identity.
Mirrors the workaround the tiny.place plugin already uses (regenerating
slash-free wallets); here we remap deterministically instead since the
identity must stay a pure function of the wallet key.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
📝 WalkthroughWalkthroughSolana tiny.place signer seed generation now remaps seeds whose derived Ed25519 public-key Base64 contains ChangesSolana identity seed handling
Possibly related issues
Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/openhuman/wallet/chains/solana.rs`:
- Around line 733-744: Add a golden-vector assertion to the slash-bearing seed
test around slash_free_identity_seed, asserting the exact expected output seed
or derived public key for the fixed slashed input; retain the existing
slash-free and determinism checks to prevent accidental identity migration from
changes to hashing or encoding.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: eb23741a-fd38-4538-a1c0-daeca88b22f8
📒 Files selected for processing (1)
src/openhuman/wallet/chains/solana.rs
| // Slashed seed is remapped to a slash-free, deterministic result. | ||
| let out = slash_free_identity_seed(slashed); | ||
| assert_ne!(out, slashed, "slashed seed must be remapped"); | ||
| assert!( | ||
| !pub_b64(&out).contains('/'), | ||
| "result pubkey must be slash-free" | ||
| ); | ||
| assert_eq!( | ||
| slash_free_identity_seed(slashed), | ||
| out, | ||
| "must be deterministic" | ||
| ); |
There was a problem hiding this comment.
🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win
Pin a golden remapping vector to prevent accidental identity migration.
Calling the helper twice only proves current-run determinism. Changes to the domain, counter encoding, or hash inputs would still pass while silently changing TinyPlace and Signal identities. Assert the exact output seed or public key for a fixed slash-bearing seed.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@src/openhuman/wallet/chains/solana.rs` around lines 733 - 744, Add a
golden-vector assertion to the slash-bearing seed test around
slash_free_identity_seed, asserting the exact expected output seed or derived
public key for the fixed slashed input; retain the existing slash-free and
determinism checks to prevent accidental identity migration from changes to
hashing or encoding.
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 010c36441f
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| // Never logged: the log below omits the seed. | ||
| log::debug!("[tinyplace] signer seed derived (seed not logged)"); | ||
| Ok(signing_key.to_bytes()) | ||
| Ok(slash_free_identity_seed(signing_key.to_bytes())) |
There was a problem hiding this comment.
For any wallet whose raw public key base64 contains /, this now returns a seed for a different Ed25519 key, so LocalSigner.agent_id() is no longer the Solana wallet address derived by the normal wallet path. The existing tiny.place payment and registry flows still treat signer.agent_id() as the wallet owner, e.g. wallet_usdc_balance(&signer.agent_id()), x402 from, and RegisterRequest.crypto_id, while execute_prepared spends from the original Solana wallet. In that slash-bearing case users can see/authorize under an unfunded or different cryptoId, and paid actions can be rejected after the on-chain transfer; keep the spending identity tied to the wallet and fix the /keys path/messaging key separately.
Useful? React with 👍 / 👎.
1 participant
Summary
slash_free_identity_seed()at the point the seed is handed toLocalSigner.Problem
The tiny.place backend serves the Ed25519 identity public key as the Signal bundle
identityKeyin STANDARD base64 and keys its/keys/{identityKey}/…routes on that string verbatim. Standard base64's alphabet includes/, which becomes%2Fin the request path — and those key routes 404 on it. So an identity whose public key's base64 contains a/cannot publish its Signal pre-keys and cannot receive DMs.Our tiny.place identity is deterministic from the Solana wallet key, and roughly half of all Ed25519 public keys contain at least one
/in standard base64. So ~half of new wallets mint a silently-unusable tiny.place identity (bundle publish + inbound DM both 404). This is the same class of failure theplugin-tinyplaceSDK already sidesteps by regenerating slash-free wallets.Solution
tinyplace_signer_seed()now returnsslash_free_identity_seed(signing_key.to_bytes()):/do we deterministically derive a fresh seed fromSHA-256("openhuman.tinyplace.identity.slashfree.v1" ‖ seed ‖ counter), bumpingcounteruntil the resulting public key is slash-free. Same wallet always maps to the same lowest-counter slash-free identity.The tiny.place plugin regenerates slash-free wallets; here the identity must remain a pure function of the wallet key, so we remap the seed deterministically instead.
Submission Checklist
slash_free_identity_seed_cleans_slash_keys_deterministicallycovers the clean-passthrough path (counter 0, unchanged) and the slashed→remapped path (slash-free + deterministic).fn+ its unit test are fully exercised (both branches). Flagging for the coverage gate to confirm.N/A: internal key-derivation fix, no user-facing feature row.N/A, uses already-importedsha2/ed25519_dalek/base64.N/A: no release-cut surface touched.Impact
debuglog omits the seed)./, this changes that wallet's advertised tiny.place identity to a different (slash-free) key. In practice those identities were already non-functional — they 404 on bundle-publish and on inbound DMs — so there is no working relay state to preserve, and remapping is strictly a repair. But if any such identity was partially advertised (e.g. published out-of-band, printed in a UI, or saved by a peer), those references will no longer resolve after upgrade. Recommend the team confirm no persisted slash-bearing identities need an explicit migration/announcement before merge. Clean identities (counter 0) are unaffected.Related
AI Authored PR Metadata
Linear Issue
Commit & Branch
fix/tinyplace-slash-free-identityValidation Run
pnpm --filter openhuman-app format:check— N/A (noapp/changes)pnpm typecheck— N/A (no TS changes)cargo test -p openhuman --lib slash_free_identity_seed_cleans_slash_keys_deterministically→ 1 passedcargo fmt --check -- src/openhuman/wallet/chains/solana.rs→ cleanBehavior Changes
Parity Contract
Summary by CodeRabbit