Information propagation observatory. Maps coupling between physical, computational, biological, and financial systems using the 12 IG primitives as a common cross-substrate vocabulary.
44 public-API domain streams (no keys): base 1–16 (fear/greed, mempool, coingecko, onchain, tides, air quality, DONKI, seismic, Kp, HN sentiment, solar wind, lightning, wikipedia, weather…), market/macro 17–24 (options skew, yield curve, VIX, shipping, grid, night lights, GDELT…), chiral bio 25–29 (genbank, pubmed…), chiral astro 30–34, SIC-POVM fill 35–38, extraplanetary 39–44 (geomag, neutron, GOES XRS, LIGO…). Each stream thresholds to primitive alert levels 0/1/2; co-activation across streams is a B-state event (Belnap Both - dialetheic confluence).
Out-of-sample gate python -m ig_pulse.oos separates the two: invariant - native-B confluence over-disperses ~3.4× vs independence null and replicates on unseen data; not invariant - the coupling graph is contemporaneous (lead-lag below time-shuffled null in ig_pulse/braid.py) and non-stationary: a now-map, not a fixed lattice.
collect → couple → map → report (snapshots.jsonl → coupling.json → graph.json → B-state report): collect via DomainStreamAggregator; couple by Pearson cross-correlation over (stream, primitive) series (|r|≥min_r, p≤max_p); map as ASCII matrix or Graphviz DOT; report reconstructs propagation anatomy per B-state timestamp. Thresholds per stream: ig_pulse/domain_streams.py.
python -m ig_pulse.cli collect --once # one cycle
python -m ig_pulse.cli collect --interval 90 # loop
python -m ig_pulse.cli couple
python -m ig_pulse.cli map # ASCII
python -m ig_pulse.cli map --dot # DOT
python -m ig_pulse.cli report --ts 2026-06-22T00:03:25Z
python -m ig_pulse.cli report
python -m ig_pulse.oos # out-of-sample gateAlert ≥3 at 1.50× marks B-state confluence. Feeds fin3r via ./run_fin3r.sh signals.
Full 507-line version: README_backups/ig-pulse_README.md (backed up only if formerly present; see README_backups/).
μ∘δ = id