Project Moonshot, LLM Evaluation Toolkit
Is this legally binding?
Guidance. Voluntary guidance. Best practice, not obligation, until a contract or a regulator cites it.
Open-source LLM evaluation toolkit from the AI Verify Foundation combining benchmarking and red teaming to help developers and compliance teams test LLMs and LLM applications. Voluntary GUIDANCE tool; launched in open beta in May 2024.
From the source
“Project Moonshot is one of the world's first Large Language Model (LLM) Evaluation Toolkits.”
Project Moonshot page
Instrument record
- Short name
- Moonshot
- Date
- 2024-05
- Version
- Not versioned
- Cluster
- core
What this connects to
9 relations. Official relations are the ones the source documents state; anything marked GAGE analysis is our reading, not an agency's.
Operationalises1
- InstrumentAI Verify Testing Framework and ToolkitGuidanceGAGE analysis, not official
Moonshot extends the AI Verify testing approach from traditional AI to LLMs via benchmarking and red teaming
Issued by2
- AgencyAI Verify Foundation (AIVF)Guidance
- AgencyInfocomm Media Development Authority (IMDA)Guidance
Launched with IMDA (open beta, May 2024)
Cites4
- ToolMoonshot (Software)Guidance
Structural decomposition of the source instrument
- ConceptBenchmarking ('Exam Questions' for LLMs)Guidance
Structural decomposition of the source instrument
- ConceptRed Teaming (Adversarial Prompting)Guidance
Structural decomposition of the source instrument
- ConceptStarter Kit for LLM-Based App TestingGuidance
Structural decomposition of the source instrument
Cited by2
- InstrumentGlobal AI Assurance Pilot (Feb 2025)Guidance
Starter Kit testing guidance draws on Global AI Assurance Pilot learnings; recommended tests are incorporated into Moonshot
- EventEnhanced Testing Framework for GenAI (29 May 2025)Guidance
2025 framework update addresses GenAI risks; Moonshot is the Foundation's companion LLM evaluation toolkit
Verified against the official source on 2026-08-17. GAGE is not affiliated with or endorsed by any agency named here, and nothing on this page is legal advice. How this is built and checked.