cohort_members
The full cohort, one record per analog, with rich per-member metadata (forward outcomes, regime, anchor fundamentals, news, chart events). Slice and bucket it yourself.
How to use it
cohort_members is exposed by the Chart Library MCP MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the cohort_members tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.
Install Chart Library MCP
claude mcp add chart-library -- chartlibrary-mcpOther tools in Chart Library MCP (15)
Analytic metrics. metric= accepts anomaly, volumeprofile, crowding, correlationshift, earningsreaction, patterndegradation, regimeaccuracy, decompose (slice winners vs losers), clusters (cohort-internal grouping).
Same engine as pullcomps under the original field names (cohortid, featureimportance, winrate, volregime, …). Kept callable verbatim for existing integrations; new ones should prefer pullcomps.
Within-cohort winner/loser attribution — which member traits separated the forward-return tail from the rest, each with a by-date cluster-bootstrap CI and a false-discovery decision. Descriptive, never causal.
Partition the cohort by one dimension (volregime, sectoretf, momentum5d, …) → per-bucket outcome distributions vs baseline. The one-call "does this dimension matter?" primitive.
Slice/probe a stored comp set by ANY attribute (macro · technical · event) and get per-subset stats vs the full-cohort baseline. No kNN re-run. "Of the 300 analogs, how do the post-earnings-week ones do?"
Reorder the cohort by a weighted composite of member fields you name (e.g. "ret5d:1,distance:-0.5") — impose your objective on the analogs, fully auditable.
Situational data. target= accepts "market", a ticker symbol ("NVDA"), {"symbol": ..., "date": ...} for lightweight anchor metadata, or "system" for DB coverage.
Narrative + rankings derived from a cohort. style= accepts filterranking (which filter shifts the distribution most), prose (plain-English summary), positionguidance (exit signals), riskranking.
portfolio
Multi-holding weighted conditional distribution. Runs per-holding cohorts in parallel, weight-averages the distributions, ranks tail contributors.
The flagship. Pull the comp set for a subject (symbol, date, timeframe) — the historical analogs, what they did next, the drivers that separated the best outcomes, and our coveragerecord. Front-of-house lexicon: subject · compsetid · compcount · compstrength · matchquality · drivers · uprate · condi
File an error or improvement suggestion back to the project.
Entry point. Find similar historical patterns for an anchor; returns a comp-set handle you can chain. mode= supports text (default), livebars (raw OHLCV), similar (cohort-level neighbors).
Layer 5 memory — per-symbol feature reliability + achieved calibration across prior analyses. Ground a read in whether a feature has historically been reliable for this ticker.
Historical predicted-vs-realized coverage of our calibrated bands (a track record, not a forecast). The nominal 80% band held 80.8% across 302,880 prior cases.