nex-code 0.4.15 → 0.4.17
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/README.md +42 -43
- package/dist/nex-code.js +433 -430
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -19,7 +19,7 @@
|
|
|
19
19
|
<img src="https://img.shields.io/badge/Ollama_Cloud-supported-brightgreen.svg" alt="Ollama Cloud: supported">
|
|
20
20
|
<img src="https://img.shields.io/badge/node-%3E%3D18-brightgreen.svg" alt="Node >= 18">
|
|
21
21
|
<img src="https://img.shields.io/badge/dependencies-2-green.svg" alt="Dependencies: 2">
|
|
22
|
-
<img src="https://img.shields.io/badge/tests-
|
|
22
|
+
<img src="https://img.shields.io/badge/tests-3453-blue.svg" alt="Tests: 3453">
|
|
23
23
|
<img src="https://img.shields.io/badge/VS_Code-extension-007ACC.svg" alt="VS Code extension">
|
|
24
24
|
</p>
|
|
25
25
|
|
|
@@ -107,7 +107,7 @@ npm update -g nex-code
|
|
|
107
107
|
|
|
108
108
|
**Extensible.** Plugin API (`registerTool` + lifecycle hooks), skill system (install from any git URL), MCP server support.
|
|
109
109
|
|
|
110
|
-
**Tested.**
|
|
110
|
+
**Tested.** 3453 tests, 79% coverage, CI on every push.
|
|
111
111
|
|
|
112
112
|
---
|
|
113
113
|
|
|
@@ -119,19 +119,18 @@ Rankings are based on nex-code's own `/benchmark` — 15 tool-calling tasks agai
|
|
|
119
119
|
### Flat-Rate / Pay-as-you-go
|
|
120
120
|
|
|
121
121
|
<!-- nex-benchmark-start -->
|
|
122
|
-
<!-- Updated: 2026-03-
|
|
122
|
+
<!-- Updated: 2026-03-26 — run `/benchmark --discover` after new Ollama Cloud releases -->
|
|
123
123
|
|
|
124
|
-
| Rank | Model
|
|
125
|
-
|
|
126
|
-
| 🥇
|
|
127
|
-
| 🥈
|
|
128
|
-
| 🥉
|
|
129
|
-
| —
|
|
130
|
-
| —
|
|
124
|
+
| Rank | Model | Score | Avg Latency | Context | Best For |
|
|
125
|
+
|---|---|---|---|---|---|
|
|
126
|
+
| 🥇 | `devstral-2:123b` | **82.5** | 1.7s | 131K | Default — fastest + most reliable tool selection |
|
|
127
|
+
| 🥈 | `devstral-small-2:24b` | 75 | 3.1s | 131K | Fast sub-agents, simple lookups |
|
|
128
|
+
| 🥉 | `qwen3-coder:480b` | 72.5 | 8.4s | 131K | Coding-heavy sessions, heavy sub-agents |
|
|
129
|
+
| — | `kimi-k2:1t` | 67.5 | 6.6s | 256K | Large repos (>100K tokens) |
|
|
130
|
+
| — | `minimax-m2.7:cloud` | 64.1 | 5.0s | 200K | Complex swarm / multi-agent sessions (Toolathon SOTA) |
|
|
131
131
|
|
|
132
132
|
> Rankings are nex-code-specific: tool name accuracy, argument validity, schema compliance.
|
|
133
133
|
> Toolathon (Minimax SOTA) measures different task types — run `/benchmark --discover` after model releases.
|
|
134
|
-
|
|
135
134
|
<!-- nex-benchmark-end -->
|
|
136
135
|
|
|
137
136
|
### Recommended `.env` for Ollama Cloud (Flat-Rate)
|
|
@@ -332,20 +331,20 @@ nex-code --prompt-file /tmp/task.txt --yolo --json
|
|
|
332
331
|
# → {"success":true,"response":"..."}
|
|
333
332
|
```
|
|
334
333
|
|
|
335
|
-
| Flag
|
|
336
|
-
|
|
|
337
|
-
| `--task <prompt>`
|
|
338
|
-
| `--prompt-file <path>`
|
|
339
|
-
| `--delete-prompt-file`
|
|
340
|
-
| `--auto`
|
|
341
|
-
| `--yolo`
|
|
342
|
-
| `--server`
|
|
343
|
-
| `--json`
|
|
344
|
-
| `--max-turns <n>`
|
|
345
|
-
| `--model <spec>`
|
|
346
|
-
| `--debug`
|
|
347
|
-
| `--auto-orchestrate`
|
|
348
|
-
| `--orchestrator-model <m>` | Model for decomposition/synthesis step (default: `kimi-k2.5`)
|
|
334
|
+
| Flag | Description |
|
|
335
|
+
| -------------------------- | ------------------------------------------------------------------------------------------------------------- |
|
|
336
|
+
| `--task <prompt>` | Run a single prompt and exit |
|
|
337
|
+
| `--prompt-file <path>` | Read prompt from a UTF-8 file and run headless |
|
|
338
|
+
| `--delete-prompt-file` | Delete the prompt file after reading (use with `--prompt-file`) |
|
|
339
|
+
| `--auto` | Skip confirmations (non-interactive, no REPL banner) |
|
|
340
|
+
| `--yolo` | Skip all confirmations including dangerous commands (also configurable via `.nex/config.json` `"yolo": true`) |
|
|
341
|
+
| `--server` | Start JSON-lines IPC server (used by the VS Code extension) |
|
|
342
|
+
| `--json` | Output `{"success":true,"response":"..."}` to stdout |
|
|
343
|
+
| `--max-turns <n>` | Override the agentic loop iteration limit |
|
|
344
|
+
| `--model <spec>` | Use a specific model (e.g. `anthropic:claude-sonnet-4-6`) |
|
|
345
|
+
| `--debug` | Show internal diagnostic messages (compression, loop detection, guards) |
|
|
346
|
+
| `--no-auto-orchestrate` | Disable auto-orchestration for multi-goal prompts (on by default; also: `NEX_AUTO_ORCHESTRATE=false`) |
|
|
347
|
+
| `--orchestrator-model <m>` | Model for decomposition/synthesis step (default: `kimi-k2.5`) |
|
|
349
348
|
|
|
350
349
|
---
|
|
351
350
|
|
|
@@ -965,19 +964,19 @@ Spawn parallel sub-agents for independent tasks:
|
|
|
965
964
|
|
|
966
965
|
For complex tasks with multiple independent goals (e.g. "fix all TypeScript errors in auth/, add tests for utils/, and update the README"), the orchestrator decomposes the prompt into parallel sub-tasks, runs dedicated sub-agents on each, and synthesizes the results.
|
|
967
966
|
|
|
968
|
-
**
|
|
967
|
+
**Auto-orchestration is on by default** for prompts with ≥3 goals.
|
|
969
968
|
|
|
970
969
|
```bash
|
|
971
|
-
#
|
|
972
|
-
nex-code --
|
|
970
|
+
# Just use it — multi-goal prompts auto-decompose into parallel agents
|
|
971
|
+
nex-code --task "fix all type errors in src/, add JSDoc to utils/, update CHANGELOG"
|
|
973
972
|
|
|
974
|
-
#
|
|
975
|
-
nex-code --
|
|
973
|
+
# Custom orchestrator model
|
|
974
|
+
nex-code --orchestrator-model kimi-k2.5 --task "..."
|
|
976
975
|
|
|
977
|
-
#
|
|
978
|
-
NEX_AUTO_ORCHESTRATE=
|
|
976
|
+
# Disable auto-orchestration
|
|
977
|
+
NEX_AUTO_ORCHESTRATE=false nex-code
|
|
979
978
|
|
|
980
|
-
#
|
|
979
|
+
# Lower the goal threshold (default: 3)
|
|
981
980
|
NEX_ORCHESTRATE_THRESHOLD=2 nex-code
|
|
982
981
|
```
|
|
983
982
|
|
|
@@ -1011,17 +1010,17 @@ Suggested commit: fix: resolve auth type errors and add utility docs
|
|
|
1011
1010
|
|
|
1012
1011
|
**Env vars:**
|
|
1013
1012
|
|
|
1014
|
-
| Variable
|
|
1015
|
-
|
|
|
1016
|
-
| `NEX_AUTO_ORCHESTRATE`
|
|
1017
|
-
| `NEX_ORCHESTRATE_THRESHOLD` | `3`
|
|
1013
|
+
| Variable | Default | Description |
|
|
1014
|
+
| --------------------------- | ------- | ------------------------------------------------------- |
|
|
1015
|
+
| `NEX_AUTO_ORCHESTRATE` | `true` | Set to `false` to disable auto-orchestration |
|
|
1016
|
+
| `NEX_ORCHESTRATE_THRESHOLD` | `3` | Minimum number of detected goals before auto-triggering |
|
|
1018
1017
|
|
|
1019
1018
|
**Model roles in orchestration:**
|
|
1020
1019
|
|
|
1021
|
-
| Role
|
|
1022
|
-
|
|
|
1023
|
-
| Orchestrator
|
|
1024
|
-
| Worker
|
|
1020
|
+
| Role | Default model | Purpose |
|
|
1021
|
+
| ------------ | ----------------- | ------------------------------------------- |
|
|
1022
|
+
| Orchestrator | `kimi-k2.5` | Decomposes prompt, synthesizes results |
|
|
1023
|
+
| Worker | `devstral-2:123b` | Executes each sub-task (one agent per task) |
|
|
1025
1024
|
|
|
1026
1025
|
Override via `--orchestrator-model` (orchestrator) or `DEFAULT_MODEL` / `NEX_STANDARD_MODEL` (workers).
|
|
1027
1026
|
|
|
@@ -1554,9 +1553,9 @@ npm test # Run all tests with coverage
|
|
|
1554
1553
|
npm run test:watch # Watch mode
|
|
1555
1554
|
```
|
|
1556
1555
|
|
|
1557
|
-
|
|
1556
|
+
83 test suites, 3453 tests, 79% statement / 71% branch coverage.
|
|
1558
1557
|
|
|
1559
|
-
CI runs on GitHub Actions (Node
|
|
1558
|
+
CI runs on GitHub Actions (Node 20 LTS).
|
|
1560
1559
|
|
|
1561
1560
|
---
|
|
1562
1561
|
|