@nexus-cortex/cli 4.116.0 → 4.117.0
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.
- package/.env.defaults +43 -0
- package/package.json +2 -2
package/.env.defaults
CHANGED
|
@@ -1010,6 +1010,49 @@ CORTEX_ENDTURN_GATE=true
|
|
|
1010
1010
|
#CORTEX_FINISH_CONFIRM=
|
|
1011
1011
|
#CORTEX_FINISH_CONFIRM_MIN_REMAINING=
|
|
1012
1012
|
#CORTEX_FINISH_CONFIRM_MAX=
|
|
1013
|
+
# CORTEX_MEETS_CONFIRM / CORTEX_MEETS_CONFIRM_MIN_REMAINING — R168 HB-MEETS-CONFIRM (4.117.0, 2026-09-17, pilot-12 rerun read). A MEETS was
|
|
1014
|
+
# unconditional: decideVetoAction accepted it on the verdict string alone, the judge's own CHECK: lines were never run for a MEETS, and
|
|
1015
|
+
# R167's hold excluded MEETS by construction — 2–3 sessions per arm finished on a WRONG MEETS at 81–92% budget left and walked straight
|
|
1016
|
+
# through (html-js-filter, mvcc-lsm-compaction, gsea). FIX: with the lever on the judge persona demands one to three `CHECK:` lines whose
|
|
1017
|
+
# PASSING proves the task's own criteria; the harness runs them at once; a MEETS with no passing evidence (nothing named, or every named
|
|
1018
|
+
# check failed/inconclusive, and the workspace's own check entry point did not pass) with >= MIN_REMAINING (0.5) of the budget left is
|
|
1019
|
+
# HELD once — sharing CORTEX_FINISH_CONFIRM_MAX with R167 — with the check results, the budget and the junior's open_items; an
|
|
1020
|
+
# unverified MEETS also no longer silences the R151/R157 nudges. A MEETS backed by a passing check stands. Event: finish_confirm
|
|
1021
|
+
# {action:'accept', meets:true, evidencePassed, checksNamed, namedRanNow, namedPassed, …}. =false restores 4.116.2 byte-for-byte.
|
|
1022
|
+
# [LEDGER 2026-09-17 @4.117.0] BUILT + unit-tested (endTurnResolver.test.ts R168). Field: PENDING (pilot-12 rerun, lever arm vs control).
|
|
1023
|
+
#CORTEX_MEETS_CONFIRM=
|
|
1024
|
+
#CORTEX_MEETS_CONFIRM_MIN_REMAINING=
|
|
1025
|
+
# CORTEX_TURN_CONTRACT_ENFORCE / CORTEX_TURN_CONTRACT_ENFORCE_MAX — HB-TURN-CONTRACT-ENFORCE (4.117.0, 2026-09-17). The channel contract as
|
|
1026
|
+
# a PROMPT clause did not take (pilot-12: 6% of tool-calling turns carried ANALYSIS + PLAN; 8/10 sessions ignored it). Terminus 2 does
|
|
1027
|
+
# not ask — its parser REJECTS a response missing the fields and re-prompts. This is that structural half: with ENFORCE=true a
|
|
1028
|
+
# tool-calling response whose visible text lacks a line-leading ANALYSIS and PLAN is rejected — its tool calls come back UNEXECUTED as
|
|
1029
|
+
# error results carrying the re-prompt — up to ENFORCE_MAX (2) times per turn, then the batch runs as-is (liveness). Finish batches
|
|
1030
|
+
# (EndTurn) are exempt. Pair with CORTEX_TURN_CONTRACT=channel (the prompt half). Events: turn_contract_reject {iteration, rejects,
|
|
1031
|
+
# max, missing, tools}, turn_contract_fallback. Default off = byte-identical.
|
|
1032
|
+
# [LEDGER 2026-09-17 @4.117.0] BUILT + unit-tested (turnContractValidator.test.ts). Field: PENDING (pilot-12 rerun, lever arm vs control).
|
|
1033
|
+
#CORTEX_TURN_CONTRACT_ENFORCE=
|
|
1034
|
+
#CORTEX_TURN_CONTRACT_ENFORCE_MAX=
|
|
1035
|
+
# CORTEX_JUDGE_TOOL_ROUNDS / CORTEX_JUDGE_TOOL_ROUND_BUDGET_MS — R170 HB-JUDGE-TOOL-LOOP (4.117.0, 2026-09-17). The v2 of
|
|
1036
|
+
# docs/LIFT_MENTOR_PLANNER_EXPERIMENT_SPEC.md §2.2/§2.4 ("bounded tool-using mentor, turn-capped never token-capped"), built as a
|
|
1037
|
+
# HARNESS-DRIVEN text protocol over the existing single-shot helper call: the helper request path carries no tools (no tools/
|
|
1038
|
+
# tool_choice/tool role in any helper adapter), but the harness already RUNS judge-named checks (R166 runCheck) — the missing edge was
|
|
1039
|
+
# that the judge never saw the results of the commands it named in the same round. With ROUNDS > 1 the persona offers
|
|
1040
|
+
# `VERDICT: INVESTIGATE` + `CHECK: <cmd>` (read-only denylist: no rm/mv/git checkout/redirect-to-file/installs) + `READ: <path>[:a-b]`
|
|
1041
|
+
# (in-workspace file slices, 120 lines / 4 KB, binary + oversize refused); the harness executes them, appends an EVIDENCE block and
|
|
1042
|
+
# asks again, withdrawing the option on the last round or when the aggregate ROUND_BUDGET_MS (240 s) or 2x that of the wall budget is
|
|
1043
|
+
# gone; an INVESTIGATE after withdrawal abstains (never re-loops). Investigation checks that pass count as evidence for R168.
|
|
1044
|
+
# The thinking-on escalation is skipped when >1 round ran. Event endturn_resolver gains {toolRounds, roundsUsed, investigateChecks,
|
|
1045
|
+
# investigateReads, investigateRefused, roundLatencyMs, evidenceChars}; `mentor` describes the LAST call. Default 1 = 4.116.2.
|
|
1046
|
+
# [LEDGER 2026-09-17 @4.117.0] BUILT + unit-tested (endTurnResolver.test.ts R170, judgeEvidence.test.ts R170). Field: PENDING (pilot-12 rerun, ROUNDS=3 arm vs control).
|
|
1047
|
+
#CORTEX_JUDGE_TOOL_ROUNDS=
|
|
1048
|
+
#CORTEX_JUDGE_TOOL_ROUND_BUDGET_MS=
|
|
1049
|
+
# (no lever) R169 HB-NEARDUP-SCRATCH-STEM — REFUTED before shipping (2026-09-17, pilot-12 trajectory replay). The distiller's 16
|
|
1050
|
+
# "repeated-identical-retry" candidates were re-read: the two stuck sessions (coq-block-bound base arm: `cat > /tmp/tN.v <<EOF` with a
|
|
1051
|
+
# DIFFERENT proof body each attempt; rs-archive-clone: 191 distinct `python3 -c` bodies) are unconverged EXPLORATION, not retries —
|
|
1052
|
+
# R136's inline-body Dice correctly keeps them apart, and replaying the similarity lens over the real commands gives max window
|
|
1053
|
+
# counts of 5 and 3 (diversify fires at 7) with OR without numbered-scratch-file stem grouping; grouping only pushed a healthy
|
|
1054
|
+
# exploration (gsea's diagN.py) from 2 to 6. The in-place case (persist coq, one Main.v) reached 15 and DID block. No identity-grouping
|
|
1055
|
+
# lever shipped; the class is a distiller artifact + hard tasks. Ledger r-tb4-pilot12-2026-09-17.md §9.
|
|
1013
1056
|
# (no lever) 4.110.0 closes the 4.109.0 disclosures (2026-09-16): R154 image-rejection heal now also covers the STREAMING loop
|
|
1014
1057
|
# (continuation re-streamed once after the stub; stream empty-retry and stream R29a synthesis wrapped like the non-streaming calls);
|
|
1015
1058
|
# the sub-agent console fallback prints a progress beat only on a turn change or every 30 s (was every second: 500+ ticker lines per
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@nexus-cortex/cli",
|
|
3
|
-
"version": "4.
|
|
3
|
+
"version": "4.117.0",
|
|
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.
|
|
22
|
+
"@nexus-cortex/core": "4.117.0",
|
|
23
23
|
"chalk": "^4.1.2",
|
|
24
24
|
"cli-spinners": "^2.9.0",
|
|
25
25
|
"commander": "^11.0.0",
|