Intelligence → Decision → Execution
A strategic intelligence ecosystem designed to turn fragmented information into explainable decisions and executable actions.
The execution gap
Information is abundant. Insight is increasingly accessible. Execution remains fragmented — it’s the gap between knowing and doing. InsightValues is being built to close that gap.
IVE
IVE is InsightValues’ reasoning and execution engine — the component that turns context into verifiable action.
IVE Strategic Execution Agent
Prototype Deployed in Production
Our submission to the All Things Agentic 2026 hackathon: an agent built with Google ADK and Gemini, deployed on Google Cloud Run, that executes authenticated Supabase tools to analyze real strategic context and create a verified action.
Google ADK
Google Cloud Run
Supabase + RLS
JWT (ES256)
IVE is deployed on Google Cloud Run and executes authenticated Supabase tools. User identity comes from the JWT, and database access is protected by Row Level Security. IVE creates actions and verifies persistence with a read-after-write check, and executions are traceable via execution_id. Verified strategic decisions can be persisted to durable business memory and retrieved as context in future executions.
create_action(…) → inserts into action_queue (RLS)
get_action(action_id) → confirms persistence (read-after-write)
write_agent_memory(…) → verified decision → business_memory
InsightValues Systems
InsightValues Platform
Strategic and business intelligence platform (Flutter + Supabase), predating the hackathon.
IVE
Intelligence and execution engine — context, reasoning, action, verification, and memory. Deployed on Google Cloud Run.
IVE Strategic Execution Agent
Reference implementation of IVE, built for the All Things Agentic 2026 hackathon — with real authentication, real persistence, and real verification.
InsightValues Quant
Quantitative research and market intelligence — in planning.
System evidence
No marketing screens — the real flow of an authenticated execution in production.
↓
validate_supabase_jwt() → ES256 / JWKS → authenticated_user_id
↓
get_project_context() → real project via Supabase (RLS)
↓
create_action(…) → inserts into action_queue (RLS)
↓
get_action(action_id) → found: true (read-after-write confirms persistence)
↓
write_agent_memory(…) → verified decision persisted to business_memory
↓
RESULT_VERIFIED
An automated test suite covers JWT/JWKS authentication, the Supabase tools contract, idempotency, verification, and the HTTP service. The write to business_memory is correct by code review; its execution in production has not yet been exercised by a dedicated smoke test.
Latest articles
InsightValues: From Intelligence to Execution
Why more data and more AI don’t automatically mean better decisions — and how we’re closing that gap.
Building IVE: A Strategic Execution Agent with Google ADK and Gemini
Architecture, tools, tests, and an honest look at the limitations of the agent built for the hackathon.
Explore the ecosystem
Early Access
Follow the preparation of the InsightValues commercial MVP and receive relevant future information when appropriate.
173 container Caixas de metadados
174 container
175 button (expanded) Caixas de metadados, Secondary Actions: Collapse
176 splitter Description: Arraste para redimensionar, Value: 77, ID: base-ui-:rb: