Dex – Cost-aware analytics engineering skills for agents
Cost guardrails for Snowflake queries when agents token-max and burn your budget.
Dex is the agent-native analytics engineering toolkit. Point it at your warehouse and your dbt project. It learns the landscape, authors your transformations, and tells you exactly what to fix when the schema drifts. Built for analytics engineers and data engineers who want more out of their coding agent.
Closes the context gap for coding agents by owning the dbt project as a first-class object.
Analytics engineers, data engineers
dbt Cloud · Metabase · Datafold
Cost guardrails for Snowflake queries when agents token-max and burn your budget.
Token burner satire with TUI dashboard, but intentionally useless by design.
Faithful end-to-end replays let you watch agents hand off work like a movie, not just see final scores.
Clickable project and organisation pages plus CSV downloads make this an immediately usable explorer for anyone auditing open-source climate work. The dashboard surfaces rankings, timelines and 'Good First Issues' links (via ClimateTriage), which is practical; however citation coverage is thin and there's limited methodological detail, so it feels like a strong MVP rather than a breakthrough.
If you run personal agents this nails a frequently ignored need: it auto-collects agent lifecycle events via Claude hooks and turns raw calls into readable daily markdown, searchable timelines with animated replay, and 30-day cost/failure analytics. The local-only model and the whimsical 'AI Journal' entries are neat differentiators — main caveat: it’s tightly coupled to Claude hooks and the install/edit-settings friction will limit casual adoption.
Every insight re-queried before shipping — catches hallucinations early.