Please do not report security vulnerabilities through public GitHub issues, discussions, or pull requests.
Send reports to security@aethelred.io with:
- affected asset, branch, tag, endpoint, or image
- issue summary and impact
- reproducible steps or proof of concept
- exploit prerequisites and assumptions
- logs, traces, or transaction references where relevant
- suggested remediation, if available
If you need an encrypted channel, email first and request a secure handoff.
Aethelred currently operates a private, invitation-only protocol bug bounty for the protocol and validator surface.
Current status:
- Protocol bug bounty: active, private by invitation
- Public protocol bug bounty: not yet open
- dApps and ecosystem repos: out of scope until separately announced
| Function | Contact |
|---|---|
| Official intake mailbox | security@aethelred.io |
| Interim triage owner | Ramesh Tamilselvan <rameshtamilselvan@gmail.com> |
| Interim reward approval owner | Ramesh Tamilselvan <rameshtamilselvan@gmail.com> |
| Interim disclosure approval owner | Ramesh Tamilselvan <rameshtamilselvan@gmail.com> |
The canonical program documents are:
- docs/security/BUG_BOUNTY_PROGRAM.md
- docs/security/BUG_BOUNTY_SEVERITY_MATRIX.md
- docs/security/BUG_BOUNTY_SLA.md
- docs/security/BUG_BOUNTY_TRIAGE_PLAYBOOK.md
- docs/security/BUG_BOUNTY_ROLLOUT_PLAN.md
In scope for the current private program:
- protocol consensus safety and liveness
- validator, staking, slashing, and governance controls
- PoUW job integrity, randomness, and settlement correctness
- TEE attestation validation and enclave trust binding
- bridge and custody-critical paths when deployed or reachable in the current release
- protocol RPC, REST, gRPC, and validator-facing security flaws
- protocol SDK issues that create a real protocol exploit path
- validator image and release artifacts in the active supported release
Out of scope unless explicitly approved in writing:
- marketing sites, brochure pages, or general website content issues
- social engineering, phishing, spam, or physical attacks
- denial-of-service without a concrete protocol exploit path
- missing headers, generic best-practice findings, or clickjacking with no material impact
- third-party library CVEs without a demonstrated Aethelred exploit path
- vulnerabilities in dApps or ecosystem repos that have not been explicitly added to scope
| Surface | Status |
|---|---|
release/testnet-v1.0 |
Supported |
main |
Supported |
ghcr.io/aethelred-foundation/aethelred/aethelredd:testnet-v1.0.1 |
Supported |
| Severity | Initial acknowledgement | Triage target | Update cadence |
|---|---|---|---|
| Critical | 24 hours | 72 hours | every 24 hours |
| High | 24 hours | 5 business days | every 3 business days |
| Medium | 48 hours | 7 business days | weekly |
| Low | 5 business days | 10 business days | at milestones |
Detailed remediation and disclosure SLAs are maintained in docs/security/BUG_BOUNTY_SLA.md.
The current protocol bounty is denominated in USD value, not in AETHEL tokens.
Default payout method:
- USDC
Fallback payout methods may be used where needed for legal, compliance, or counterparty reasons.
Severity bands and payout ranges are defined in docs/security/BUG_BOUNTY_SEVERITY_MATRIX.md.
If you act in good faith, avoid privacy violations and destructive testing, and follow this policy, Aethelred will not pursue action against compliant security research performed within scope.
Researchers must:
- avoid accessing, modifying, or exfiltrating third-party data
- avoid degrading availability for real users, validators, or counterparties
- stop testing and report immediately once sensitive data or systemic access is obtained
- avoid persistence, backdoors, or post-exploitation activity beyond what is needed to prove impact
- Audit evidence index: docs/audits/README.md
- Audit status tracker: docs/audits/STATUS.md
- Release provenance: docs/security/release-provenance.md
- Fail closed: security-critical paths reject on unexpected state
- Deterministic consensus: no floating point or nondeterministic arithmetic in consensus
- Attestation binding: enclave trust is bound to chain and workload context
- Defense in depth: compile-time, genesis-time, and runtime gates are all used
- Operational traceability: security-relevant changes require evidence and provenance