ts-util-core 2.0.2 → 2.0.4
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 +11 -1
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -85,6 +85,10 @@ await VIEW.load(container, { url: '/api/partial-view' });
|
|
|
85
85
|
| "Did you validate the form?" | Validation is automatic — can't skip |
|
|
86
86
|
| Code review debates on style | Architecture enforces the style |
|
|
87
87
|
|
|
88
|
+
- **Eliminates divergence**: Engineers learn one API — no debates over implementation details.
|
|
89
|
+
- **Enforces consistency**: Every request passes through the same pipeline, ensuring loading overlays aren't missed and validation isn't skipped.
|
|
90
|
+
- **Reduces onboarding cost**: A new team member reads one `AJAX.request()` example and is ready to go, instead of deciphering ten scattered patterns.
|
|
91
|
+
|
|
88
92
|
### For AI Agents
|
|
89
93
|
|
|
90
94
|
| Before | After |
|
|
@@ -95,13 +99,19 @@ await VIEW.load(container, { url: '/api/partial-view' });
|
|
|
95
99
|
| Must audit every Agent's output for missing steps | Pipeline guarantees completeness — **guardrail by design** |
|
|
96
100
|
| Agent "forgets" loading overlay | Impossible — architecture enforces it |
|
|
97
101
|
|
|
102
|
+
When multiple AI Agents co-produce code, this abstraction layer becomes even more critical:
|
|
103
|
+
|
|
104
|
+
- **Token efficiency**: An Agent only needs to emit `AJAX.request({ url, form })` — one line — instead of expanding the full `fetch` + validation + error handling + loading logic every time. Context window is AI's most precious resource; saving tokens means preserving quality.
|
|
105
|
+
- **Predictable behavior**: Code generated by different Agents flows through the same pipeline, guaranteeing consistent results. You don't need to audit whether each Agent correctly implemented the loading overlay.
|
|
106
|
+
- **Guardrail effect**: The abstraction layer itself acts as a guardrail. An Agent cannot "forget" to validate a form because `AJAX.request()` does it automatically. Discipline is enforced by architecture, not memory.
|
|
107
|
+
|
|
98
108
|
### The Core Insight
|
|
99
109
|
|
|
100
110
|
> **Discipline is not "remembering to do the right thing." Discipline is making the right thing the only thing that can happen.**
|
|
101
111
|
>
|
|
102
112
|
> That's what TS-Util does — for your team today, and for the AI Agents that will write most of your code tomorrow.
|
|
103
113
|
|
|
104
|
-
📖 Read
|
|
114
|
+
📖 Read this in other languages: [繁體中文](https://github.com/MattAtAIEra/TS-Util/blob/main/docs/Readme-translations_zh.md) · [日本語](https://github.com/MattAtAIEra/TS-Util/blob/main/docs/Readme-translations_jp.md) · [한국어](https://github.com/MattAtAIEra/TS-Util/blob/main/docs/Readme-translations_ko.md) · [Español](https://github.com/MattAtAIEra/TS-Util/blob/main/docs/Readme-translations_es.md) · [Deutsch](https://github.com/MattAtAIEra/TS-Util/blob/main/docs/Readme-translations_de.md)
|
|
105
115
|
|
|
106
116
|
---
|
|
107
117
|
|