dsh-agy 0.3.1 → 0.4.0
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/README.md +29 -5
- package/lib/{accounts-CkaHhb5H.mjs → accounts-BN4h0AU4.mjs} +446 -63
- package/lib/accounts-BN4h0AU4.mjs.map +1 -0
- package/lib/cli/index.mjs +5 -5
- package/lib/client.js +25 -3
- package/lib/{constants-Db4tmyfb.mjs → constants-BpyRirvk.mjs} +38 -2
- package/lib/constants-BpyRirvk.mjs.map +1 -0
- package/lib/{exchange-yZPIZumk.mjs → exchange-LzEyxve1.mjs} +2 -2
- package/lib/{exchange-yZPIZumk.mjs.map → exchange-LzEyxve1.mjs.map} +1 -1
- package/lib/{import-BaFULQnH.mjs → import-CotRHNDk.mjs} +3 -3
- package/lib/{import-BaFULQnH.mjs.map → import-CotRHNDk.mjs.map} +1 -1
- package/lib/index.mjs +2 -2
- package/lib/{models-BGuwl50d.mjs → models-Qn0-sh6X.mjs} +2 -2
- package/lib/{models-BGuwl50d.mjs.map → models-Qn0-sh6X.mjs.map} +1 -1
- package/lib/{plugin-common-B1MG00JT.mjs → plugin-common-N7p9Cm-3.mjs} +58 -14
- package/lib/plugin-common-N7p9Cm-3.mjs.map +1 -0
- package/lib/{quota-summary-5CypVffd.mjs → quota-summary-adID82I2.mjs} +9 -12
- package/lib/quota-summary-adID82I2.mjs.map +1 -0
- package/lib/{translate-BQu8HLvB.mjs → translate-YhGDUuVa.mjs} +157 -23
- package/lib/translate-YhGDUuVa.mjs.map +1 -0
- package/lib/web/plugin.d.mts +28 -1
- package/lib/web/plugin.d.mts.map +1 -1
- package/lib/web/plugin.mjs +35 -19
- package/lib/web/plugin.mjs.map +1 -1
- package/package.json +9 -4
- package/lib/accounts-CkaHhb5H.mjs.map +0 -1
- package/lib/constants-Db4tmyfb.mjs.map +0 -1
- package/lib/plugin-common-B1MG00JT.mjs.map +0 -1
- package/lib/quota-summary-5CypVffd.mjs.map +0 -1
- package/lib/translate-BQu8HLvB.mjs.map +0 -1
package/README.md
CHANGED
|
@@ -20,8 +20,8 @@ fingerprinting, and both CLI and web management.
|
|
|
20
20
|
rate limits, per-account cooldown to the real reset time, per-account device
|
|
21
21
|
fingerprints.
|
|
22
22
|
- **Inline Settings UI**: an Antigravity section inside DSH Settings with four
|
|
23
|
-
tabs — accounts (login, activation,
|
|
24
|
-
fingerprint and proxy management), models (per-model visibility), usage
|
|
23
|
+
tabs — accounts (login, activation, grouped 5-hour/weekly quota windows, test
|
|
24
|
+
calls, fingerprint and proxy management), models (per-model visibility), usage
|
|
25
25
|
(cumulative token and request statistics), and credentials (import/export).
|
|
26
26
|
No separate page: the surface only exists where DSH Settings does.
|
|
27
27
|
- **Model visibility toggles**: hide individual models from the DSH model
|
|
@@ -37,10 +37,14 @@ fingerprinting, and both CLI and web management.
|
|
|
37
37
|
|
|
38
38
|
## Screenshots
|
|
39
39
|
|
|
40
|
-
The Antigravity section inside DSH Settings
|
|
41
|
-
|
|
40
|
+
The Antigravity section inside DSH Settings. Account emails and project names are
|
|
41
|
+
redacted.
|
|
42
42
|
|
|
43
|
-
|
|
43
|
+
| Accounts | Models |
|
|
44
|
+
|:---:|:---:|
|
|
45
|
+
|  |  |
|
|
46
|
+
| **Limits** — 5-hour and weekly windows | **Usage** — cumulative, per-model and per-account |
|
|
47
|
+
|  |  |
|
|
44
48
|
|
|
45
49
|
## Quickstart
|
|
46
50
|
|
|
@@ -193,6 +197,26 @@ reset with headroom left are used first ("use it or lose it"), near-exhausted
|
|
|
193
197
|
families are avoided, and exhausted families block the account until the real
|
|
194
198
|
reset time.
|
|
195
199
|
|
|
200
|
+
Each family is measured on **both** of its windows, because they refill on
|
|
201
|
+
different clocks:
|
|
202
|
+
|
|
203
|
+
| Window | Refills | Exhausted when | Blocks until |
|
|
204
|
+
|---|---|---|---|
|
|
205
|
+
| Rolling 5-hour | every 5 hours | below 15% left | the 5-hour reset |
|
|
206
|
+
| Weekly | every 7 days | 1% or less left | the **weekly** reset |
|
|
207
|
+
|
|
208
|
+
The two gates are independent. A family can sit at 90% of its 5-hour budget
|
|
209
|
+
while its week is spent — four 5-hour refills do not return a weekly budget —
|
|
210
|
+
so a week-exhausted account is rotated away instead of being selected and
|
|
211
|
+
failing. The thresholds are deliberately different (15% of five hours is ~45
|
|
212
|
+
minutes of runway; 1% of a week is ~1.7 hours): judging the 5-hour window too
|
|
213
|
+
late costs a wait of at most five hours, while judging the week too late parks
|
|
214
|
+
the account for days.
|
|
215
|
+
|
|
216
|
+
The weekly reading comes only from `retrieveUserQuotaSummary` — the per-model
|
|
217
|
+
`fetchAvailableModels` probe has no window field at all — and a probe that
|
|
218
|
+
reports neither window cannot erase a known-drained one.
|
|
219
|
+
|
|
196
220
|
429 (Too Many Requests) responses:
|
|
197
221
|
|
|
198
222
|
| Category | Behavior |
|