@nexus-cortex/cli 4.124.40 → 4.124.41

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/.env.defaults +28 -0
  2. package/package.json +2 -2
package/.env.defaults CHANGED
@@ -1003,6 +1003,34 @@ CORTEX_ENDTURN_GATE=true
1003
1003
  # per orchestrator. Event pid1_guard {noted, pid1Cmd, children, chars} when the note is added. off: nothing read, nothing added.
1004
1004
  # [LEDGER 2026-10-01 @4.124.40] DARK — BUILT + unit-tested (pid1Guard.test.ts: fake /proc builder cases). NO bench arm yet.
1005
1005
  #CORTEX_PID1_GUARD=
1006
+ # CORTEX_PLATEAU_STOP / _REL / _WINDOW / _MIN_POINTS — R210 HB-PLATEAU-STOP (DARK, 2026-10-02). v39a layout-config-recreation2: the
1007
+ # agent cannot see the grader's pass threshold on an optimization task, so it polished a self-computed score for 5.3 h (483 iterations,
1008
+ # 840,636 → 851,547 with ever-finer searches); budget steering reports time left, never "enough". Operator design = a PLATEAU VARIATION
1009
+ # TARGET. on: at the first tool_result boundary (same carrier as the R209 PID 1 note) ONE system-reminder asks the model, when optimizing
1010
+ # a measurable objective, to print `PLATEAU_METRIC <name>=<number> dir=<max|min>` after every evaluation and keep the deliverable at its
1011
+ # best result. Every tool round, the tool_result texts are scanned for that line (last occurrence per result wins); a per-session series
1012
+ # is kept per metric name (orchestrator/plateauStop.ts). Plateau = n >= MIN_POINTS and the best of the last WINDOW reports improves on
1013
+ # the best before them by < REL, relative (max: (bestRecent-bestBefore)/|bestBefore|; min inverted; a series that got worse counts).
1014
+ # The first plateau steers ONE reminder on the budget-steering channel (appended to the round's first tool_result, steering_injected
1015
+ # kind 'plateau'): "PLATEAU: your <name> improved only <x>% over the last <window> evaluations (target ≥ <rel>%). Stop optimizing.
1016
+ # Make sure the deliverable holds the best result (<best>), run the task's own verification/tests once, then finish." At most 2 per
1017
+ # session; the second only after >= WINDOW more reports with still no real gain. Events: plateau_metric_seen {name, dir, value} (once
1018
+ # per metric name), plateau_stop {name, dir, best, improvementPct, points, firing}. STATUS segment (with CORTEX_TURN_STATUS on): once a
1019
+ # metric was reported, the per-turn STATUS line gains ONE segment for the most recently reported metric — `score <name>: best <v> (<n>
1020
+ # evals); +<x>% over last <window> (target ≥<rel>%)`, plus ` — PLATEAU` while the detector says plateau (improvement part only once the
1021
+ # series is comparable). No metric seen or lever off → STATUS byte-identical. off: no instruction, results not scanned, no segment.
1022
+ # REL = relative-improvement target (default 0.002 = 0.2%); WINDOW = most recent reports compared (default 6); MIN_POINTS = reports
1023
+ # needed before any stop can fire (default 8). Bad values fall back to the defaults.
1024
+ # [LEDGER 2026-10-02 @4.124.41] DARK — BUILT + unit-tested (plateauStop.test.ts: detector rising/flat/worse/min/min-points/window edges,
1025
+ # noisy-line parsing, firing cap, lever-off path, STATUS segment; plateauStop.integration.test.ts: both loops, off = no instruction/
1026
+ # reminder/segment, on = instruction + 2 reminders + PLATEAU segment). NO bench arm yet. evidence: R210 v39a layout-config-recreation2 (483 iterations, 5.3 h).
1027
+ #CORTEX_PLATEAU_STOP=
1028
+ # [LEDGER 2026-10-02 @4.124.41] DARK — default 0.002 (0.2%) relative; untested on a bench. evidence: R210 v39a layout (840,636 → 851,547 over 5.3 h; per-window gains not yet measured).
1029
+ #CORTEX_PLATEAU_REL=
1030
+ # [LEDGER 2026-10-02 @4.124.41] DARK — default 6 reports; untested on a bench. evidence: R210 v39a layout.
1031
+ #CORTEX_PLATEAU_WINDOW=
1032
+ # [LEDGER 2026-10-02 @4.124.41] DARK — default 8 reports; untested on a bench. evidence: R210 v39a layout.
1033
+ #CORTEX_PLATEAU_MIN_POINTS=
1006
1034
  # (no lever) R154 HB-IMAGE-REJECTION-HEAL (4.109.0, 2026-09-16): retro-console-soc generated a 128x128 8-bit GRAYSCALE PNG, called
1007
1035
  # ReadImage, and DeepSeek answered the next continuation with `400 .messages[290].image[0]: You have uploaded an unsupported image`;
1008
1036
  # the exception escaped after the tool batch, the already-executed guard left the loop, and the R29a synthesis failed on the same block —
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@nexus-cortex/cli",
3
- "version": "4.124.40",
3
+ "version": "4.124.41",
4
4
  "description": "Nexus Cortex CLI - Terminal interface for multi-provider LLM orchestration",
5
5
  "main": "dist/index.js",
6
6
  "type": "module",
@@ -19,7 +19,7 @@
19
19
  "prepack": "node ../../scripts/copy-pkg-cortex-scaffold.mjs"
20
20
  },
21
21
  "dependencies": {
22
- "@nexus-cortex/core": "4.124.40",
22
+ "@nexus-cortex/core": "4.124.41",
23
23
  "chalk": "^4.1.2",
24
24
  "cli-spinners": "^2.9.0",
25
25
  "commander": "^11.0.0",