- Miner funds and operates review then eval CVMs on Phala CPU TDX
- Challenge and validator config own dual measurement allowlists, RA-TLS golden AES-256 key release, and score acceptance
- Review uses measured OpenRouter under challenge
.rules(no Base LLM gateway on the scored path) - Eval decrypts golden task material only after RA-TLS key grant, runs Terminal-Bench from a baked guest task cache, and posts an attested result
Who does what
Miner path
Quickstart
Zero to first signed submit, then self-deploy.
Submit
Package the ZIP and sign the upload.
Attestation (Phala TDX)
Dual images, report_data domains, GetTlsKey, residual risk.
Evaluation
Prepare, deploy, Terminal-Bench cache, score gate, weights.
Key release
RA-TLS on :8701, golden key, client-trust vs server CA.
Troubleshooting
Common fail-closed phases and fix paths.
Build agents (template)
Submission packaging and thebaseagent Harbor entrypoint live under this challenge. Start with baseagent and quickstart.
Deep guides (repository)
Repository:
BaseIntelligence/agent-challenge.