Repository navigation
Commit 5eec9b0
authored
fix: restore pending frontend HITL tools after replay (#7271)
Reloading an Intelligence thread restored frontend HITL tool calls but
left them without an active response handler, so the user could see the
approval UI without being able to answer it.
This adds an internal frontend-tool classification and tags HITL
registrations in React, Vue, and Angular. Shared core restores
unanswered HITL handlers after replay and continues the agent with the
user's result. Ordinary frontend tools remain passive; native interrupt
handling is unchanged. React Native and legacy React wrappers inherit
the React registration.
Restoration finishes independently of the connection lifecycle,
preserves sequential prompts across reconnects, ignores saved or
remotely supplied answers, and cancels superseded interactions when
switching threads. The classification is not sent in the AG-UI tool
schema.
Validation: formatting and lint pass; explicit type checks, full test
suites, and builds pass for core, React, Vue, and Angular. Coverage
includes replay → answer → continuation, saved and remote answers,
repeated reconnects, sequential prompts, thread switching, and keeping
tool classification off the wire. Commit hooks also passed tests and
package-export checks across 24 affected projects. Live Intelligence
browser validation remains a follow-up before marking this draft ready.
PNI-449
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Bug Fixes**
* Improved human-in-the-loop behavior after reconnecting or switching
conversations.
* Pending approval requests now resume correctly without duplicating
completed or remotely answered requests.
* Multi-step responses remain sequential, while stale responses are
safely ignored.
* Regular tools continue operating normally during standard agent runs.
* Stopping one agent no longer interrupts another agent’s pending
interaction.
* Cancelled interactions now report handler errors correctly when
explicitly stopped.
* **Improvements**
* Human-in-the-loop tools are explicitly distinguished from regular
tools across Angular, React, and Vue.
* Pending interactions can be cancelled safely using abort signals.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->16 files changed
Lines changed: 1120 additions & 175 deletions
File tree
- packages
- angular/src/lib
- directives/__tests__
- core/src
- __tests__
- core
- react-core/src/v2
- hooks
- __tests__
- providers
- __tests__
- vue/src/v2
- hooks
- __tests__
- providers
- __tests__
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
267 | 267 | | |
268 | 268 | | |
269 | 269 | | |
270 | | - | |
| 270 | + | |
271 | 271 | | |
272 | 272 | | |
273 | 273 | | |
| |||
277 | 277 | | |
278 | 278 | | |
279 | 279 | | |
280 | | - | |
| 280 | + | |
281 | 281 | | |
282 | 282 | | |
283 | 283 | | |
| 284 | + | |
284 | 285 | | |
285 | 286 | | |
286 | 287 | | |
287 | 288 | | |
288 | 289 | | |
289 | 290 | | |
| 291 | + | |
290 | 292 | | |
291 | 293 | | |
292 | | - | |
| 294 | + | |
| 295 | + | |
| 296 | + | |
| 297 | + | |
| 298 | + | |
293 | 299 | | |
294 | 300 | | |
295 | 301 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
567 | 567 | | |
568 | 568 | | |
569 | 569 | | |
570 | | - | |
571 | | - | |
| 570 | + | |
| 571 | + | |
| 572 | + | |
| 573 | + | |
| 574 | + | |
| 575 | + | |
| 576 | + | |
572 | 577 | | |
573 | 578 | | |
574 | 579 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
1 | 1 | | |
2 | | - | |
| 2 | + | |
3 | 3 | | |
4 | 4 | | |
5 | 5 | | |
6 | | - | |
| 6 | + | |
| 7 | + | |
7 | 8 | | |
8 | | - | |
| 9 | + | |
| 10 | + | |
9 | 11 | | |
| 12 | + | |
10 | 13 | | |
11 | 14 | | |
12 | 15 | | |
13 | 16 | | |
14 | 17 | | |
15 | 18 | | |
16 | 19 | | |
17 | | - | |
18 | | - | |
19 | | - | |
20 | 20 | | |
21 | 21 | | |
22 | 22 | | |
| |||
33 | 33 | | |
34 | 34 | | |
35 | 35 | | |
36 | | - | |
37 | | - | |
38 | | - | |
39 | 36 | | |
40 | 37 | | |
41 | 38 | | |
| |||
48 | 45 | | |
49 | 46 | | |
50 | 47 | | |
| 48 | + | |
| 49 | + | |
| 50 | + | |
| 51 | + | |
| 52 | + | |
| 53 | + | |
| 54 | + | |
| 55 | + | |
| 56 | + | |
| 57 | + | |
| 58 | + | |
| 59 | + | |
| 60 | + | |
| 61 | + | |
| 62 | + | |
| 63 | + | |
| 64 | + | |
| 65 | + | |
| 66 | + | |
| 67 | + | |
| 68 | + | |
| 69 | + | |
| 70 | + | |
| 71 | + | |
| 72 | + | |
| 73 | + | |
| 74 | + | |
| 75 | + | |
| 76 | + | |
| 77 | + | |
| 78 | + | |
| 79 | + | |
| 80 | + | |
| 81 | + | |
| 82 | + | |
| 83 | + | |
| 84 | + | |
| 85 | + | |
| 86 | + | |
| 87 | + | |
| 88 | + | |
| 89 | + | |
| 90 | + | |
| 91 | + | |
| 92 | + | |
| 93 | + | |
| 94 | + | |
| 95 | + | |
| 96 | + | |
| 97 | + | |
51 | 98 | | |
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
1 | 1 | | |
2 | | - | |
| 2 | + | |
| 3 | + | |
| 4 | + | |
| 5 | + | |
| 6 | + | |
| 7 | + | |
| 8 | + | |
| 9 | + | |
| 10 | + | |
3 | 11 | | |
4 | 12 | | |
5 | 13 | | |
| |||
13 | 21 | | |
14 | 22 | | |
15 | 23 | | |
16 | | - | |
17 | | - | |
18 | | - | |
19 | | - | |
20 | | - | |
21 | | - | |
22 | | - | |
23 | | - | |
24 | | - | |
25 | | - | |
26 | | - | |
27 | | - | |
| 24 | + | |
| 25 | + | |
| 26 | + | |
| 27 | + | |
| 28 | + | |
| 29 | + | |
| 30 | + | |
| 31 | + | |
| 32 | + | |
28 | 33 | | |
| 34 | + | |
| 35 | + | |
| 36 | + | |
| 37 | + | |
| 38 | + | |
29 | 39 | | |
| 40 | + | |
| 41 | + | |
| 42 | + | |
| 43 | + | |
| 44 | + | |
| 45 | + | |
| 46 | + | |
| 47 | + | |
| 48 | + | |
| 49 | + | |
| 50 | + | |
| 51 | + | |
| 52 | + | |
| 53 | + | |
| 54 | + | |
| 55 | + | |
30 | 56 | | |
31 | 57 | | |
0 commit comments