v3.64.0 — verified-outcome learning loop

Minor release: @claude-flow/cli, claude-flow, ruflo 3.64.0; ruflo-mods plugin 0.8.0.

Added

• Verified-outcome learning loop (#4581, #4583, ADR-502), on by default (ruflo-mods learningLoop; set it to false to turn it off).
  • The mod records each main-loop turn. Its outcome is decided only by the exit status of a test, build, typecheck or lint command run after the turn's last edit.
  • || true, pipes and ; verify nothing, and the model's own words are never read.
  • ruflo hooks verified-ingest, the MCP tool hooks_intelligence_verified-ingest, and automatic ingest in hooks_route and hooks_session-end all train SONA on verified turns only. Unverified turns are counted and never trained.
  • Learned lessons at or above the 0.6 routing gate are written to .claude-flow/data/learned-lessons.json. The mod cites at most three of them in a prompt they match.
  • ruflo doctor --component learning-loop reports whether outcomes are reaching SONA.

Not yet

• #4581 stages 5–7 (console pane, autonomous /loop-run, GEPA self-tuning) are not built.
• The #4581 §7 evaluation (30+ tasks, baseline vs loop-enabled) has not been run; the default is on by maintainer decision ahead of it (ADR-502 §6).

Install: npx ruflo@3.64.0 (also @claude-flow/cli and claude-flow 3.64.0; latest/alpha/v3alpha all point here).

The ruflo-mods half (0.8.0) ships through the ruflo plugin marketplace.