* docs: update HERALD.md — S63-S64, standards 32, fresh-eyes CLI results Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * feat(system): S65 night shift — fix 14 branches from fresh-eyes CLI testing (FPLAN-0154) 14 branches dispatched across 3 phases, all fixes verified. Phase 1 (critical): backup dry-run flag check, ai_mail per-ID branch detection, flow introspection gate removal, drone atomic writes + empty file guard, commons 3-strategy caller detection. Phase 2 (patterns): spawn argparse --help + create --dry-run, skills create --help, api introspection routing fix (6 commands), trigger empty JSON guard, seedgo @ prefix consistency + conftest. Phase 3 (polish): prax ghost module archived + real status output, daemon update digest + 13 error cascade fixes + branch-health all-branches, memory search timeout message + hidden modules in help, cli drone_adapter archived + standalone verified. 3,410 tests (+80), seedgo 99% avg, 0 failures. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Tests
Pytest unit tests for MEMORY.
conftest.py— Shared fixtures (temp dirs, mocks, sample data).test_*.py— Test files. Standard tests cover JSON handler, CLI routing, and error resilience. Custom tests cover branch-specific domain logic.