Love what OpenAlice is — a genuine research → entry → management → exit loop in one agent is rare.

The one primitive I keep finding missing in that loop: at entry, a grounded base rate for what analogous setups actually did next (most agents confidently narrate what a chart "should" do — the "Oracle Fallacy"). Chart Library is an MCP node that returns, for a (ticker, date), the cohort of real historical analogs and a calibrated forward-return distribution. Historical, never a prediction — the 80% band held 80.8% across 302,880 cases.

Small open-source reference crew using it as an optional node, with a with/without answer-lift comparison: https://github.com/grahammccain/chart-library-agent-crew

Happy to open an off-by-default PR if you'd want to try it at the entry step — no worries if not.

— Graham, chartlibrary.io