CleverCon currently runs on Stellar Testnet. Contracts, wallets, and funds involved are all testnet assets with no real-world value. Even so, we treat security issues seriously, since the CleverVault contract, the PolicyVerifier, and the non-custodial delegate + settlement path are the foundation for a future mainnet deployment.

If you discover a security vulnerability, please report it privately by emailing joshuaibitoye111@gmail.com. Do not open a public GitHub issue for security reports.

Please include:

• A description of the vulnerability and its potential impact.
• Steps to reproduce, including any relevant transaction hashes, contract IDs, or request payloads (testnet only; do not send real secret keys).
• The affected package or contract (e.g. contracts/agent-vault, services/api).

• Acknowledgement within 48 hours of your report.
• Initial triage within 5 business days, including a severity assessment and, where applicable, a plan and rough timeline for a fix.
• We keep you updated as a fix is developed and let you know when it ships. Credit is offered to reporters who wish to be named once a fix is released.

In scope:

• contracts/agent-vault (CleverVault), contracts/policy-verifier, and contracts/registry: fund-handling logic, authorization checks, the proof-gated release + verifier cross-call, and state transitions.
• services/api, services/workers, services/indexer, and packages/{common,db}: SEP-10/JWT auth, scoped API keys, step-up auth, the delegate that signs vault calls, the settlement path, and how the delegate secret is stored (AES-256-GCM encrypted in Postgres).
• packages/agent-sdk and packages/mcp: the spending surfaces and how a scoped API key is bounded by policy.
• The private spending-policy layer (proof verification, policy commitments, and replay/nullifier handling), including the circuits/spend-policy Noir circuit.
• scripts/* and CI/deployment configuration (the Dockerfile, .github/workflows/*).

Out of scope:

• General UI/UX issues in apps/web. Please still report these, but they are a lower priority during the current backend-hardening phase.
• Third-party services CleverCon depends on (the Stellar network itself, the x402 facilitator, Anthropic's API, the host platform). Report these to their respective maintainers.
• Issues that require a compromised local environment or physical access to a user's machine.

We state these plainly rather than hide them:

• Privacy is v1. The on-chain PolicyVerifier performs a host-accelerated binding check over the public inputs, not full pairing-based verification (Soroban has no pairing host function), so the proving stack is trusted for predicate soundness. In v1 the caps/allowlist are also enforced off-chain by the API, with the on-chain budget + commitment + binding proof as the anchor. Full on-chain verification is pending Stellar pairing precompiles. See docs/private-policies.md section 7.
• No third-party audit yet. The contracts and API have not had an external security audit; that is a pre-mainnet roadmap item.
• Delegate keys. The per-user delegate secret is stored AES-256-GCM encrypted in Postgres; the encryption key (DELEGATE_ENCRYPTION_KEY) is an operator responsibility (a KMS is the mainnet plan).

If you find additional issues along these lines, please still report them. Duplicates help us prioritize.