copilotkit 4.12.0 → 4.13.1
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 +10 -0
- package/cli-build-info.json +7 -7
- package/index.js +1590 -348
- package/onboarding/index.json +20 -2
- package/onboarding/prompts/authenticate/start.md +38 -33
- package/onboarding/prompts/conversion/plan.md +3 -3
- package/onboarding/prompts/credentials/finalize-plan.md +9 -8
- package/onboarding/prompts/credentials/plan.md +26 -20
- package/onboarding/prompts/credentials/settle-credentials.md +8 -8
- package/onboarding/prompts/credentials/write-plan.md +19 -6
- package/onboarding/prompts/fallback/best-effort.md +6 -6
- package/onboarding/prompts/feature/a2ui/implement.md +6 -6
- package/onboarding/prompts/feature/a2ui/proof.md +6 -6
- package/onboarding/prompts/feature/a2ui/start.md +7 -7
- package/onboarding/prompts/feature/channels/implement.md +75 -12
- package/onboarding/prompts/feature/channels/proof.md +13 -6
- package/onboarding/prompts/feature/channels/start.md +27 -10
- package/onboarding/prompts/feature/chat-suggestions/implement.md +6 -6
- package/onboarding/prompts/feature/chat-suggestions/proof.md +6 -6
- package/onboarding/prompts/feature/chat-suggestions/start.md +7 -7
- package/onboarding/prompts/feature/complete.md +1 -1
- package/onboarding/prompts/feature/learning/implement.md +15 -15
- package/onboarding/prompts/feature/learning/proof.md +7 -7
- package/onboarding/prompts/feature/learning/start.md +8 -8
- package/onboarding/prompts/feature/open-generative-ui/implement.md +6 -6
- package/onboarding/prompts/feature/open-generative-ui/proof.md +6 -6
- package/onboarding/prompts/feature/open-generative-ui/start.md +7 -7
- package/onboarding/prompts/feature/realtime-sync/implement.md +7 -7
- package/onboarding/prompts/feature/realtime-sync/proof.md +6 -6
- package/onboarding/prompts/feature/realtime-sync/start.md +6 -6
- package/onboarding/prompts/feature/rich-threads/implement.md +8 -8
- package/onboarding/prompts/feature/rich-threads/proof.md +6 -6
- package/onboarding/prompts/feature/rich-threads/start.md +6 -6
- package/onboarding/prompts/feature/stop.md +2 -2
- package/onboarding/prompts/feature/voice/implement.md +6 -6
- package/onboarding/prompts/feature/voice/proof.md +6 -6
- package/onboarding/prompts/feature/voice/start.md +7 -7
- package/onboarding/prompts/framework/ag2.md +2 -2
- package/onboarding/prompts/framework/agno.md +2 -2
- package/onboarding/prompts/framework/built-in.md +4 -2
- package/onboarding/prompts/framework/claude-sdk-python.md +2 -2
- package/onboarding/prompts/framework/claude-sdk-typescript.md +2 -2
- package/onboarding/prompts/framework/crewai-flows.md +2 -2
- package/onboarding/prompts/framework/deep-agents.md +5 -2
- package/onboarding/prompts/framework/google-adk.md +4 -2
- package/onboarding/prompts/framework/langgraph-fastapi.md +4 -3
- package/onboarding/prompts/framework/langgraph-python.md +6 -2
- package/onboarding/prompts/framework/langgraph-typescript.md +6 -2
- package/onboarding/prompts/framework/llamaindex.md +2 -2
- package/onboarding/prompts/framework/mastra.md +4 -2
- package/onboarding/prompts/framework/ms-agent-dotnet.md +4 -2
- package/onboarding/prompts/framework/ms-agent-harness-dotnet.md +2 -2
- package/onboarding/prompts/framework/ms-agent-python.md +4 -2
- package/onboarding/prompts/framework/pydantic-ai.md +2 -2
- package/onboarding/prompts/framework/strands-python.md +4 -2
- package/onboarding/prompts/framework/strands-typescript.md +4 -2
- package/onboarding/prompts/frontend/angular.md +3 -3
- package/onboarding/prompts/frontend/nextjs.md +3 -3
- package/onboarding/prompts/frontend/plan.md +21 -10
- package/onboarding/prompts/frontend/react-native.md +2 -2
- package/onboarding/prompts/frontend/react-spa.md +2 -2
- package/onboarding/prompts/frontend/vue.md +2 -2
- package/onboarding/prompts/implementation/build-and-validate.md +33 -15
- package/onboarding/prompts/proof/complete.md +35 -16
- package/onboarding/prompts/proof/oss-baseline.md +5 -5
- package/onboarding/prompts/proof/round-trip.md +23 -12
- package/onboarding/prompts/research/gather.md +12 -102
- package/onboarding/prompts/research/merge.md +60 -0
- package/onboarding/prompts/research/preflight.md +75 -0
- package/onboarding/prompts/research/route.md +30 -6
- package/onboarding/prompts/starter/clone.md +5 -5
- package/onboarding/prompts/stopped/run-failed.md +2 -2
- package/onboarding/prompts/subagent/create-plan.md +18 -1
- package/onboarding/prompts/subagent/implement-and-validate.md +37 -0
- package/onboarding/prompts/subagent/inspect-repository.md +1 -1
- package/onboarding/prompts/subagent/prove-oss-baseline.md +1 -1
- package/onboarding/prompts/subagent/prove-round-trip.md +17 -7
- package/onboarding/prompts/unsupported/no-validated-path.md +2 -2
- package/package.json +4 -1
- package/release/release-tool.js +101 -1
|
@@ -7,7 +7,7 @@ Container. Confirm the thread remains associated with the expected user.
|
|
|
7
7
|
One command decides it:
|
|
8
8
|
|
|
9
9
|
```text
|
|
10
|
-
npx --yes copilotkit@4.
|
|
10
|
+
npx --yes copilotkit@4.13.1 verify --expect-runtime intelligence --round-trip --agent <agent-id> --expect-learning-container <id> --json
|
|
11
11
|
```
|
|
12
12
|
|
|
13
13
|
It reads the thread back from the platform, so it answers for any runtime mount, and it
|
|
@@ -32,24 +32,24 @@ same threshold. Existing threads do not join automatically just because a contai
|
|
|
32
32
|
Report each attempt at the proof as it ends, counting from one:
|
|
33
33
|
|
|
34
34
|
```text
|
|
35
|
-
npx --yes copilotkit@4.
|
|
35
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase journey-attempted --attempt 1
|
|
36
36
|
```
|
|
37
37
|
|
|
38
38
|
Report each repair cycle the same way, counting from one:
|
|
39
39
|
|
|
40
40
|
```text
|
|
41
|
-
npx --yes copilotkit@4.
|
|
41
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase repair-attempted --attempt 1
|
|
42
42
|
```
|
|
43
43
|
|
|
44
44
|
After the final attempt, report the gate exactly once:
|
|
45
45
|
|
|
46
46
|
```text
|
|
47
|
-
npx --yes copilotkit@4.
|
|
47
|
+
npx --yes copilotkit@4.13.1 onboard proof --step round-trip --outcome <passed|failed|skipped>
|
|
48
48
|
```
|
|
49
49
|
|
|
50
50
|
Use `passed` only for a proved Container assignment, `failed` for an attempted proof that
|
|
51
51
|
failed, and `skipped` when the proof could not run. Then run
|
|
52
|
-
`npx --yes copilotkit@4.
|
|
52
|
+
`npx --yes copilotkit@4.13.1 onboard audit`. Continue only when it starts with
|
|
53
53
|
`Status: passed`.
|
|
54
54
|
|
|
55
55
|
If the audit fails, never repair, reset, or revert a protected path. Compare each named path
|
|
@@ -64,9 +64,9 @@ Run the audit again after each accepted or authorized change.
|
|
|
64
64
|
If the audit still fails, or starts with `Status: blocked`, route out and stop:
|
|
65
65
|
|
|
66
66
|
```text
|
|
67
|
-
npx --yes copilotkit@4.
|
|
67
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
68
68
|
```
|
|
69
69
|
|
|
70
70
|
When the audit passes, run
|
|
71
|
-
`npx --yes copilotkit@4.
|
|
71
|
+
`npx --yes copilotkit@4.13.1 onboard read feature/complete` with the actual surface
|
|
72
72
|
outcome.
|
|
@@ -11,7 +11,7 @@ friction command without another developer question. Do not ask the developer ab
|
|
|
11
11
|
telemetry: the command applies the setting they already have.
|
|
12
12
|
|
|
13
13
|
```text
|
|
14
|
-
npx --yes copilotkit@4.
|
|
14
|
+
npx --yes copilotkit@4.13.1 onboard friction --phase stop --category <slug>
|
|
15
15
|
```
|
|
16
16
|
|
|
17
17
|
Write one or two sentences to standard input: the step you stopped at and what stopped it.
|
|
@@ -38,7 +38,7 @@ status:
|
|
|
38
38
|
before the capture, the inspection, and any edit:
|
|
39
39
|
|
|
40
40
|
```text
|
|
41
|
-
npx --yes copilotkit@4.
|
|
41
|
+
npx --yes copilotkit@4.13.1 onboard read feature/blocked-by-plan
|
|
42
42
|
```
|
|
43
43
|
|
|
44
44
|
- `LEARNING_AVAILABILITY_UNAVAILABLE` means the platform did not resolve the answer. It is
|
|
@@ -59,7 +59,7 @@ again: it has no answer, so it carries the question forward rather than ending o
|
|
|
59
59
|
Before any subagent or project process runs, capture the developer's existing work:
|
|
60
60
|
|
|
61
61
|
```text
|
|
62
|
-
npx --yes copilotkit@4.
|
|
62
|
+
npx --yes copilotkit@4.13.1 onboard protect
|
|
63
63
|
```
|
|
64
64
|
|
|
65
65
|
Keep the exact protected path list it prints and give that list to every subagent. No
|
|
@@ -75,7 +75,7 @@ onboarding.
|
|
|
75
75
|
Wait for the inspection subagent to finish. Then report that the inspection came back:
|
|
76
76
|
|
|
77
77
|
```text
|
|
78
|
-
npx --yes copilotkit@4.
|
|
78
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase research-returned
|
|
79
79
|
```
|
|
80
80
|
|
|
81
81
|
A refused checkpoint prints its reason and leaves onboarding unaffected. It is not a
|
|
@@ -85,7 +85,7 @@ If the inspection did not prove the baseline this intent extends, stop here with
|
|
|
85
85
|
changing files:
|
|
86
86
|
|
|
87
87
|
```text
|
|
88
|
-
npx --yes copilotkit@4.
|
|
88
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
89
89
|
```
|
|
90
90
|
|
|
91
91
|
Fetch the current official guides before planning:
|
|
@@ -117,7 +117,7 @@ one.
|
|
|
117
117
|
After approval, record each approved path before implementation:
|
|
118
118
|
|
|
119
119
|
```text
|
|
120
|
-
npx --yes copilotkit@4.
|
|
120
|
+
npx --yes copilotkit@4.13.1 onboard protect --authorize --path <path> --reason "<the plan's sentence>"
|
|
121
121
|
```
|
|
122
122
|
|
|
123
123
|
If an approved path changed after capture and no implementation step has run, add
|
|
@@ -127,7 +127,7 @@ authorize a path the approved plan did not list.
|
|
|
127
127
|
Then report the plan this run is about to implement:
|
|
128
128
|
|
|
129
129
|
```text
|
|
130
|
-
npx --yes copilotkit@4.
|
|
130
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase plan-written
|
|
131
131
|
```
|
|
132
132
|
|
|
133
|
-
Then run `npx --yes copilotkit@4.
|
|
133
|
+
Then run `npx --yes copilotkit@4.13.1 onboard read feature/learning/implement`.
|
|
@@ -16,7 +16,7 @@ validation results without secrets.
|
|
|
16
16
|
After validation and each repair, run:
|
|
17
17
|
|
|
18
18
|
```text
|
|
19
|
-
npx --yes copilotkit@4.
|
|
19
|
+
npx --yes copilotkit@4.13.1 onboard audit
|
|
20
20
|
```
|
|
21
21
|
|
|
22
22
|
Continue only when it starts with `Status: passed`. A path under `Authorized to modify:` is
|
|
@@ -27,7 +27,7 @@ with the implementation subagent's `Files changed` section. If that section does
|
|
|
27
27
|
the path, accept the developer's external change:
|
|
28
28
|
|
|
29
29
|
```text
|
|
30
|
-
npx --yes copilotkit@4.
|
|
30
|
+
npx --yes copilotkit@4.13.1 onboard protect --accept-external --path <path>
|
|
31
31
|
```
|
|
32
32
|
|
|
33
33
|
For an env file where the developer placed a requested credential, use
|
|
@@ -36,24 +36,24 @@ or its report does not settle who changed it, ask the developer to allow the unp
|
|
|
36
36
|
change. Only after they agree, record their answer:
|
|
37
37
|
|
|
38
38
|
```text
|
|
39
|
-
npx --yes copilotkit@4.
|
|
39
|
+
npx --yes copilotkit@4.13.1 onboard protect --authorize --unplanned --path <path> --reason "<what the developer said>"
|
|
40
40
|
```
|
|
41
41
|
|
|
42
42
|
Run the audit again after each accepted or authorized change. If it still fails, or starts
|
|
43
43
|
with `Status: blocked`, route out and stop:
|
|
44
44
|
|
|
45
45
|
```text
|
|
46
|
-
npx --yes copilotkit@4.
|
|
46
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
47
47
|
```
|
|
48
48
|
|
|
49
49
|
When implementation validation passes, report it:
|
|
50
50
|
|
|
51
51
|
```text
|
|
52
|
-
npx --yes copilotkit@4.
|
|
52
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase build-validated
|
|
53
53
|
```
|
|
54
54
|
|
|
55
55
|
If validation cannot pass, or this intent needs a prerequisite the app does not have, use
|
|
56
56
|
the feature stop route above without further changes.
|
|
57
57
|
|
|
58
58
|
Otherwise run
|
|
59
|
-
`npx --yes copilotkit@4.
|
|
59
|
+
`npx --yes copilotkit@4.13.1 onboard read feature/open-generative-ui/proof`.
|
|
@@ -13,23 +13,23 @@ the missing visual proof honestly.
|
|
|
13
13
|
Report each attempt at the proof as it ends, counting from one:
|
|
14
14
|
|
|
15
15
|
```text
|
|
16
|
-
npx --yes copilotkit@4.
|
|
16
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase journey-attempted --attempt 1
|
|
17
17
|
```
|
|
18
18
|
|
|
19
19
|
Report each repair cycle the same way, counting from one:
|
|
20
20
|
|
|
21
21
|
```text
|
|
22
|
-
npx --yes copilotkit@4.
|
|
22
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase repair-attempted --attempt 1
|
|
23
23
|
```
|
|
24
24
|
|
|
25
25
|
After the final attempt, report the gate exactly once:
|
|
26
26
|
|
|
27
27
|
```text
|
|
28
|
-
npx --yes copilotkit@4.
|
|
28
|
+
npx --yes copilotkit@4.13.1 onboard proof --step round-trip --outcome <passed|failed|skipped>
|
|
29
29
|
```
|
|
30
30
|
|
|
31
31
|
Use `passed` only for a proved generated UI, `failed` for an attempted proof that failed,
|
|
32
|
-
and `skipped` when the proof could not run. Then run `npx --yes copilotkit@4.
|
|
32
|
+
and `skipped` when the proof could not run. Then run `npx --yes copilotkit@4.13.1 onboard audit`.
|
|
33
33
|
Continue only when it starts with `Status: passed`.
|
|
34
34
|
|
|
35
35
|
If the audit fails, never repair, reset, or revert a protected path. Compare each named path
|
|
@@ -44,9 +44,9 @@ Run the audit again after each accepted or authorized change.
|
|
|
44
44
|
If the audit still fails, or starts with `Status: blocked`, route out and stop:
|
|
45
45
|
|
|
46
46
|
```text
|
|
47
|
-
npx --yes copilotkit@4.
|
|
47
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
48
48
|
```
|
|
49
49
|
|
|
50
50
|
When the audit passes, run
|
|
51
|
-
`npx --yes copilotkit@4.
|
|
51
|
+
`npx --yes copilotkit@4.13.1 onboard read feature/complete` with the outcome actually
|
|
52
52
|
observed.
|
|
@@ -6,7 +6,7 @@ implementation, and proof subagents, all restricted to the target project direct
|
|
|
6
6
|
Before any subagent or project process runs, capture the developer's existing work:
|
|
7
7
|
|
|
8
8
|
```text
|
|
9
|
-
npx --yes copilotkit@4.
|
|
9
|
+
npx --yes copilotkit@4.13.1 onboard protect
|
|
10
10
|
```
|
|
11
11
|
|
|
12
12
|
Keep the exact protected path list it prints and give that list to every subagent. No
|
|
@@ -16,7 +16,7 @@ approved authorization before implementation.
|
|
|
16
16
|
Before edits, have a research subagent identify the current CopilotKit runtime route,
|
|
17
17
|
provider/chat component, agent id, package versions, existing middleware, sandbox/CSP
|
|
18
18
|
constraints, and normal test/dev commands. Prove the existing app first with `/info`,
|
|
19
|
-
`npx --yes copilotkit@4.
|
|
19
|
+
`npx --yes copilotkit@4.13.1 verify --expect-runtime oss --round-trip --agent <agent-id> --json`,
|
|
20
20
|
and a real frontend request when browser control is available. If the app is not an existing
|
|
21
21
|
OSS CopilotKit app with a proven round trip, change nothing and direct the developer to the
|
|
22
22
|
generic onboarding path first.
|
|
@@ -24,7 +24,7 @@ generic onboarding path first.
|
|
|
24
24
|
Wait for the inspection subagent to finish. Then report that the inspection came back:
|
|
25
25
|
|
|
26
26
|
```text
|
|
27
|
-
npx --yes copilotkit@4.
|
|
27
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase research-returned
|
|
28
28
|
```
|
|
29
29
|
|
|
30
30
|
A refused checkpoint prints its reason and leaves onboarding unaffected. It is not a
|
|
@@ -34,7 +34,7 @@ If the inspection did not prove the baseline this intent extends, stop here with
|
|
|
34
34
|
changing files:
|
|
35
35
|
|
|
36
36
|
```text
|
|
37
|
-
npx --yes copilotkit@4.
|
|
37
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
38
38
|
```
|
|
39
39
|
|
|
40
40
|
Do not sign in, select an Intelligence project, create credentials, change the existing
|
|
@@ -56,7 +56,7 @@ one.
|
|
|
56
56
|
After approval, record each approved path before implementation:
|
|
57
57
|
|
|
58
58
|
```text
|
|
59
|
-
npx --yes copilotkit@4.
|
|
59
|
+
npx --yes copilotkit@4.13.1 onboard protect --authorize --path <path> --reason "<the plan's sentence>"
|
|
60
60
|
```
|
|
61
61
|
|
|
62
62
|
If an approved path changed after capture and no implementation step has run, add
|
|
@@ -66,7 +66,7 @@ authorize a path the approved plan did not list.
|
|
|
66
66
|
Then report the plan this run is about to implement:
|
|
67
67
|
|
|
68
68
|
```text
|
|
69
|
-
npx --yes copilotkit@4.
|
|
69
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase plan-written
|
|
70
70
|
```
|
|
71
71
|
|
|
72
|
-
Then run `npx --yes copilotkit@4.
|
|
72
|
+
Then run `npx --yes copilotkit@4.13.1 onboard read feature/open-generative-ui/implement`.
|
|
@@ -17,14 +17,14 @@ non-zero on an app that is working. Pass what `identifyUser` reads with a repeat
|
|
|
17
17
|
It is not a defect to repair.
|
|
18
18
|
|
|
19
19
|
Run focused tests and
|
|
20
|
-
`npx --yes copilotkit@4.
|
|
20
|
+
`npx --yes copilotkit@4.13.1 verify --expect-runtime intelligence --round-trip --agent <agent-id> --json`.
|
|
21
21
|
Repair changed-file failures and record changed paths and secret-safe validation
|
|
22
22
|
evidence.
|
|
23
23
|
|
|
24
24
|
After validation and each repair, run:
|
|
25
25
|
|
|
26
26
|
```text
|
|
27
|
-
npx --yes copilotkit@4.
|
|
27
|
+
npx --yes copilotkit@4.13.1 onboard audit
|
|
28
28
|
```
|
|
29
29
|
|
|
30
30
|
Continue only when it starts with `Status: passed`. A path under `Authorized to modify:` is
|
|
@@ -35,7 +35,7 @@ with the implementation subagent's `Files changed` section. If that section does
|
|
|
35
35
|
the path, accept the developer's external change:
|
|
36
36
|
|
|
37
37
|
```text
|
|
38
|
-
npx --yes copilotkit@4.
|
|
38
|
+
npx --yes copilotkit@4.13.1 onboard protect --accept-external --path <path>
|
|
39
39
|
```
|
|
40
40
|
|
|
41
41
|
For an env file where the developer placed a requested credential, use
|
|
@@ -44,24 +44,24 @@ or its report does not settle who changed it, ask the developer to allow the unp
|
|
|
44
44
|
change. Only after they agree, record their answer:
|
|
45
45
|
|
|
46
46
|
```text
|
|
47
|
-
npx --yes copilotkit@4.
|
|
47
|
+
npx --yes copilotkit@4.13.1 onboard protect --authorize --unplanned --path <path> --reason "<what the developer said>"
|
|
48
48
|
```
|
|
49
49
|
|
|
50
50
|
Run the audit again after each accepted or authorized change. If it still fails, or starts
|
|
51
51
|
with `Status: blocked`, route out and stop:
|
|
52
52
|
|
|
53
53
|
```text
|
|
54
|
-
npx --yes copilotkit@4.
|
|
54
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
55
55
|
```
|
|
56
56
|
|
|
57
57
|
When implementation validation passes, report it:
|
|
58
58
|
|
|
59
59
|
```text
|
|
60
|
-
npx --yes copilotkit@4.
|
|
60
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase build-validated
|
|
61
61
|
```
|
|
62
62
|
|
|
63
63
|
If validation cannot pass, or this intent needs a prerequisite the app does not have, use
|
|
64
64
|
the feature stop route above without further changes.
|
|
65
65
|
|
|
66
66
|
Otherwise run
|
|
67
|
-
`npx --yes copilotkit@4.
|
|
67
|
+
`npx --yes copilotkit@4.13.1 onboard read feature/realtime-sync/proof`.
|
|
@@ -14,23 +14,23 @@ the missing proof rather than claiming realtime works.
|
|
|
14
14
|
Report each attempt at the proof as it ends, counting from one:
|
|
15
15
|
|
|
16
16
|
```text
|
|
17
|
-
npx --yes copilotkit@4.
|
|
17
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase journey-attempted --attempt 1
|
|
18
18
|
```
|
|
19
19
|
|
|
20
20
|
Report each repair cycle the same way, counting from one:
|
|
21
21
|
|
|
22
22
|
```text
|
|
23
|
-
npx --yes copilotkit@4.
|
|
23
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase repair-attempted --attempt 1
|
|
24
24
|
```
|
|
25
25
|
|
|
26
26
|
After the final attempt, report the gate exactly once:
|
|
27
27
|
|
|
28
28
|
```text
|
|
29
|
-
npx --yes copilotkit@4.
|
|
29
|
+
npx --yes copilotkit@4.13.1 onboard proof --step round-trip --outcome <passed|failed|skipped>
|
|
30
30
|
```
|
|
31
31
|
|
|
32
32
|
Use `passed` only for proved two-client sync, `failed` for an attempted proof that failed,
|
|
33
|
-
and `skipped` when the proof could not run. Then run `npx --yes copilotkit@4.
|
|
33
|
+
and `skipped` when the proof could not run. Then run `npx --yes copilotkit@4.13.1 onboard audit`.
|
|
34
34
|
Continue only when it starts with `Status: passed`.
|
|
35
35
|
|
|
36
36
|
If the audit fails, never repair, reset, or revert a protected path. Compare each named path
|
|
@@ -45,8 +45,8 @@ Run the audit again after each accepted or authorized change.
|
|
|
45
45
|
If the audit still fails, or starts with `Status: blocked`, route out and stop:
|
|
46
46
|
|
|
47
47
|
```text
|
|
48
|
-
npx --yes copilotkit@4.
|
|
48
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
49
49
|
```
|
|
50
50
|
|
|
51
51
|
When the audit passes, run
|
|
52
|
-
`npx --yes copilotkit@4.
|
|
52
|
+
`npx --yes copilotkit@4.13.1 onboard read feature/complete` with the actual outcome.
|
|
@@ -7,7 +7,7 @@ two-context proof to subagents within the target project.
|
|
|
7
7
|
Before any subagent or project process runs, capture the developer's existing work:
|
|
8
8
|
|
|
9
9
|
```text
|
|
10
|
-
npx --yes copilotkit@4.
|
|
10
|
+
npx --yes copilotkit@4.13.1 onboard protect
|
|
11
11
|
```
|
|
12
12
|
|
|
13
13
|
Keep the exact protected path list it prints and give that list to every subagent. No
|
|
@@ -23,7 +23,7 @@ unchanged and direct the developer to generic onboarding first.
|
|
|
23
23
|
Wait for the inspection subagent to finish. Then report that the inspection came back:
|
|
24
24
|
|
|
25
25
|
```text
|
|
26
|
-
npx --yes copilotkit@4.
|
|
26
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase research-returned
|
|
27
27
|
```
|
|
28
28
|
|
|
29
29
|
A refused checkpoint prints its reason and leaves onboarding unaffected. It is not a
|
|
@@ -33,7 +33,7 @@ If the inspection did not prove the baseline this intent extends, stop here with
|
|
|
33
33
|
changing files:
|
|
34
34
|
|
|
35
35
|
```text
|
|
36
|
-
npx --yes copilotkit@4.
|
|
36
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
37
37
|
```
|
|
38
38
|
|
|
39
39
|
Fetch the current official guides before planning:
|
|
@@ -59,7 +59,7 @@ one.
|
|
|
59
59
|
After approval, record each approved path before implementation:
|
|
60
60
|
|
|
61
61
|
```text
|
|
62
|
-
npx --yes copilotkit@4.
|
|
62
|
+
npx --yes copilotkit@4.13.1 onboard protect --authorize --path <path> --reason "<the plan's sentence>"
|
|
63
63
|
```
|
|
64
64
|
|
|
65
65
|
If an approved path changed after capture and no implementation step has run, add
|
|
@@ -69,7 +69,7 @@ authorize a path the approved plan did not list.
|
|
|
69
69
|
Then report the plan this run is about to implement:
|
|
70
70
|
|
|
71
71
|
```text
|
|
72
|
-
npx --yes copilotkit@4.
|
|
72
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase plan-written
|
|
73
73
|
```
|
|
74
74
|
|
|
75
|
-
Then run `npx --yes copilotkit@4.
|
|
75
|
+
Then run `npx --yes copilotkit@4.13.1 onboard read feature/realtime-sync/implement`.
|
|
@@ -2,7 +2,7 @@
|
|
|
2
2
|
|
|
3
3
|
Delegate implementation to one subagent with the approved plan and fetched official guides.
|
|
4
4
|
If repository evidence does not prove an existing valid Intelligence selection, run
|
|
5
|
-
`npx --yes copilotkit@4.
|
|
5
|
+
`npx --yes copilotkit@4.13.1 login --json` and follow its single streaming session. Then
|
|
6
6
|
list projects, show the developer the current/recent choices, and select or create one only
|
|
7
7
|
after they choose. Require the secret-safe selection summary to confirm the project file,
|
|
8
8
|
environment file, and key provisioning; never display a key. If existing configuration is
|
|
@@ -20,14 +20,14 @@ non-zero on an app that is working. Pass what `identifyUser` reads with a repeat
|
|
|
20
20
|
It is not a defect to repair.
|
|
21
21
|
|
|
22
22
|
Run focused type/test commands, then start the app and run
|
|
23
|
-
`npx --yes copilotkit@4.
|
|
23
|
+
`npx --yes copilotkit@4.13.1 verify --expect-runtime intelligence --round-trip --agent <agent-id> --json`.
|
|
24
24
|
Repair changed-file failures before continuing. Record changed paths and secret-safe
|
|
25
25
|
validation evidence.
|
|
26
26
|
|
|
27
27
|
After validation and each repair, run:
|
|
28
28
|
|
|
29
29
|
```text
|
|
30
|
-
npx --yes copilotkit@4.
|
|
30
|
+
npx --yes copilotkit@4.13.1 onboard audit
|
|
31
31
|
```
|
|
32
32
|
|
|
33
33
|
Continue only when it starts with `Status: passed`. A path under `Authorized to modify:` is
|
|
@@ -38,7 +38,7 @@ with the implementation subagent's `Files changed` section. If that section does
|
|
|
38
38
|
the path, accept the developer's external change:
|
|
39
39
|
|
|
40
40
|
```text
|
|
41
|
-
npx --yes copilotkit@4.
|
|
41
|
+
npx --yes copilotkit@4.13.1 onboard protect --accept-external --path <path>
|
|
42
42
|
```
|
|
43
43
|
|
|
44
44
|
For an env file where the developer placed a requested credential, use
|
|
@@ -47,24 +47,24 @@ or its report does not settle who changed it, ask the developer to allow the unp
|
|
|
47
47
|
change. Only after they agree, record their answer:
|
|
48
48
|
|
|
49
49
|
```text
|
|
50
|
-
npx --yes copilotkit@4.
|
|
50
|
+
npx --yes copilotkit@4.13.1 onboard protect --authorize --unplanned --path <path> --reason "<what the developer said>"
|
|
51
51
|
```
|
|
52
52
|
|
|
53
53
|
Run the audit again after each accepted or authorized change. If it still fails, or starts
|
|
54
54
|
with `Status: blocked`, route out and stop:
|
|
55
55
|
|
|
56
56
|
```text
|
|
57
|
-
npx --yes copilotkit@4.
|
|
57
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
58
58
|
```
|
|
59
59
|
|
|
60
60
|
When implementation validation passes, report it:
|
|
61
61
|
|
|
62
62
|
```text
|
|
63
|
-
npx --yes copilotkit@4.
|
|
63
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase build-validated
|
|
64
64
|
```
|
|
65
65
|
|
|
66
66
|
If validation cannot pass, or this intent needs a prerequisite the app does not have, use
|
|
67
67
|
the feature stop route above without further changes.
|
|
68
68
|
|
|
69
69
|
Otherwise run
|
|
70
|
-
`npx --yes copilotkit@4.
|
|
70
|
+
`npx --yes copilotkit@4.13.1 onboard read feature/rich-threads/proof`.
|
|
@@ -13,24 +13,24 @@ running.
|
|
|
13
13
|
Report each attempt at the proof as it ends, counting from one:
|
|
14
14
|
|
|
15
15
|
```text
|
|
16
|
-
npx --yes copilotkit@4.
|
|
16
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase journey-attempted --attempt 1
|
|
17
17
|
```
|
|
18
18
|
|
|
19
19
|
Report each repair cycle the same way, counting from one:
|
|
20
20
|
|
|
21
21
|
```text
|
|
22
|
-
npx --yes copilotkit@4.
|
|
22
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase repair-attempted --attempt 1
|
|
23
23
|
```
|
|
24
24
|
|
|
25
25
|
After the final attempt, report the gate exactly once:
|
|
26
26
|
|
|
27
27
|
```text
|
|
28
|
-
npx --yes copilotkit@4.
|
|
28
|
+
npx --yes copilotkit@4.13.1 onboard proof --step round-trip --outcome <passed|failed|skipped>
|
|
29
29
|
```
|
|
30
30
|
|
|
31
31
|
Use `passed` only for proved persistence and reopen, `failed` for an attempted proof that
|
|
32
32
|
failed, and `skipped` when the proof could not run. Then run
|
|
33
|
-
`npx --yes copilotkit@4.
|
|
33
|
+
`npx --yes copilotkit@4.13.1 onboard audit`. Continue only when it starts with
|
|
34
34
|
`Status: passed`.
|
|
35
35
|
|
|
36
36
|
If the audit fails, never repair, reset, or revert a protected path. Compare each named path
|
|
@@ -45,9 +45,9 @@ Run the audit again after each accepted or authorized change.
|
|
|
45
45
|
If the audit still fails, or starts with `Status: blocked`, route out and stop:
|
|
46
46
|
|
|
47
47
|
```text
|
|
48
|
-
npx --yes copilotkit@4.
|
|
48
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
49
49
|
```
|
|
50
50
|
|
|
51
51
|
When the audit passes, run
|
|
52
|
-
`npx --yes copilotkit@4.
|
|
52
|
+
`npx --yes copilotkit@4.13.1 onboard read feature/complete` with the actual browser
|
|
53
53
|
outcome.
|
|
@@ -7,7 +7,7 @@ implementation, and proof to separate subagents within the target project.
|
|
|
7
7
|
Before any subagent or project process runs, capture the developer's existing work:
|
|
8
8
|
|
|
9
9
|
```text
|
|
10
|
-
npx --yes copilotkit@4.
|
|
10
|
+
npx --yes copilotkit@4.13.1 onboard protect
|
|
11
11
|
```
|
|
12
12
|
|
|
13
13
|
Keep the exact protected path list it prints and give that list to every subagent. No
|
|
@@ -23,7 +23,7 @@ an existing CopilotKit app, leave it unchanged and direct the developer to gener
|
|
|
23
23
|
Wait for the inspection subagent to finish. Then report that the inspection came back:
|
|
24
24
|
|
|
25
25
|
```text
|
|
26
|
-
npx --yes copilotkit@4.
|
|
26
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase research-returned
|
|
27
27
|
```
|
|
28
28
|
|
|
29
29
|
A refused checkpoint prints its reason and leaves onboarding unaffected. It is not a
|
|
@@ -33,7 +33,7 @@ If the inspection did not prove the baseline this intent extends, stop here with
|
|
|
33
33
|
changing files:
|
|
34
34
|
|
|
35
35
|
```text
|
|
36
|
-
npx --yes copilotkit@4.
|
|
36
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
37
37
|
```
|
|
38
38
|
|
|
39
39
|
Fetch the current official guides before planning:
|
|
@@ -61,7 +61,7 @@ one.
|
|
|
61
61
|
After approval, record each approved path before implementation:
|
|
62
62
|
|
|
63
63
|
```text
|
|
64
|
-
npx --yes copilotkit@4.
|
|
64
|
+
npx --yes copilotkit@4.13.1 onboard protect --authorize --path <path> --reason "<the plan's sentence>"
|
|
65
65
|
```
|
|
66
66
|
|
|
67
67
|
If an approved path changed after capture and no implementation step has run, add
|
|
@@ -71,7 +71,7 @@ authorize a path the approved plan did not list.
|
|
|
71
71
|
Then report the plan this run is about to implement:
|
|
72
72
|
|
|
73
73
|
```text
|
|
74
|
-
npx --yes copilotkit@4.
|
|
74
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase plan-written
|
|
75
75
|
```
|
|
76
76
|
|
|
77
|
-
Then run `npx --yes copilotkit@4.
|
|
77
|
+
Then run `npx --yes copilotkit@4.13.1 onboard read feature/rich-threads/implement`.
|
|
@@ -20,7 +20,7 @@ a `start` with no `--run` reprints the id this stopped run already holds, and on
|
|
|
20
20
|
covering both reads as one run that did two different things.
|
|
21
21
|
|
|
22
22
|
```text
|
|
23
|
-
npx --yes copilotkit@4.
|
|
23
|
+
npx --yes copilotkit@4.13.1 onboard start --run <new-12-character-id>
|
|
24
24
|
```
|
|
25
25
|
|
|
26
26
|
Do not read a generic onboarding node from this run. A feature run and a generic run are
|
|
@@ -35,7 +35,7 @@ Send one short report. Run the friction command without another developer questi
|
|
|
35
35
|
ask the developer about telemetry: the command applies the setting they already have.
|
|
36
36
|
|
|
37
37
|
```text
|
|
38
|
-
npx --yes copilotkit@4.
|
|
38
|
+
npx --yes copilotkit@4.13.1 onboard friction --phase stop --category <slug>
|
|
39
39
|
```
|
|
40
40
|
|
|
41
41
|
Write one or two sentences to standard input: the step you stopped at and what stopped it.
|
|
@@ -14,7 +14,7 @@ voice proof. Record changed paths and validation results.
|
|
|
14
14
|
After validation and each repair, run:
|
|
15
15
|
|
|
16
16
|
```text
|
|
17
|
-
npx --yes copilotkit@4.
|
|
17
|
+
npx --yes copilotkit@4.13.1 onboard audit
|
|
18
18
|
```
|
|
19
19
|
|
|
20
20
|
Continue only when it starts with `Status: passed`. A path under `Authorized to modify:` is
|
|
@@ -25,7 +25,7 @@ with the implementation subagent's `Files changed` section. If that section does
|
|
|
25
25
|
the path, accept the developer's external change:
|
|
26
26
|
|
|
27
27
|
```text
|
|
28
|
-
npx --yes copilotkit@4.
|
|
28
|
+
npx --yes copilotkit@4.13.1 onboard protect --accept-external --path <path>
|
|
29
29
|
```
|
|
30
30
|
|
|
31
31
|
For an env file where the developer placed a requested credential, use
|
|
@@ -34,24 +34,24 @@ or its report does not settle who changed it, ask the developer to allow the unp
|
|
|
34
34
|
change. Only after they agree, record their answer:
|
|
35
35
|
|
|
36
36
|
```text
|
|
37
|
-
npx --yes copilotkit@4.
|
|
37
|
+
npx --yes copilotkit@4.13.1 onboard protect --authorize --unplanned --path <path> --reason "<what the developer said>"
|
|
38
38
|
```
|
|
39
39
|
|
|
40
40
|
Run the audit again after each accepted or authorized change. If it still fails, or starts
|
|
41
41
|
with `Status: blocked`, route out and stop:
|
|
42
42
|
|
|
43
43
|
```text
|
|
44
|
-
npx --yes copilotkit@4.
|
|
44
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
45
45
|
```
|
|
46
46
|
|
|
47
47
|
When implementation validation passes, report it:
|
|
48
48
|
|
|
49
49
|
```text
|
|
50
|
-
npx --yes copilotkit@4.
|
|
50
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase build-validated
|
|
51
51
|
```
|
|
52
52
|
|
|
53
53
|
If validation cannot pass, or this intent needs a prerequisite the app does not have, use
|
|
54
54
|
the feature stop route above without further changes.
|
|
55
55
|
|
|
56
56
|
Otherwise run
|
|
57
|
-
`npx --yes copilotkit@4.
|
|
57
|
+
`npx --yes copilotkit@4.13.1 onboard read feature/voice/proof`.
|
|
@@ -13,23 +13,23 @@ path, agent response, app URL, and process IDs. Fix changed-file defects and rep
|
|
|
13
13
|
Report each attempt at the proof as it ends, counting from one:
|
|
14
14
|
|
|
15
15
|
```text
|
|
16
|
-
npx --yes copilotkit@4.
|
|
16
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase journey-attempted --attempt 1
|
|
17
17
|
```
|
|
18
18
|
|
|
19
19
|
Report each repair cycle the same way, counting from one:
|
|
20
20
|
|
|
21
21
|
```text
|
|
22
|
-
npx --yes copilotkit@4.
|
|
22
|
+
npx --yes copilotkit@4.13.1 onboard checkpoint --phase repair-attempted --attempt 1
|
|
23
23
|
```
|
|
24
24
|
|
|
25
25
|
After the final attempt, report the gate exactly once:
|
|
26
26
|
|
|
27
27
|
```text
|
|
28
|
-
npx --yes copilotkit@4.
|
|
28
|
+
npx --yes copilotkit@4.13.1 onboard proof --step round-trip --outcome <passed|failed|skipped>
|
|
29
29
|
```
|
|
30
30
|
|
|
31
31
|
Use `passed` only for proved voice input, `failed` for an attempted proof that failed, and
|
|
32
|
-
`skipped` when the proof could not run. Then run `npx --yes copilotkit@4.
|
|
32
|
+
`skipped` when the proof could not run. Then run `npx --yes copilotkit@4.13.1 onboard audit`.
|
|
33
33
|
Continue only when it starts with `Status: passed`.
|
|
34
34
|
|
|
35
35
|
If the audit fails, never repair, reset, or revert a protected path. Compare each named path
|
|
@@ -44,9 +44,9 @@ Run the audit again after each accepted or authorized change.
|
|
|
44
44
|
If the audit still fails, or starts with `Status: blocked`, route out and stop:
|
|
45
45
|
|
|
46
46
|
```text
|
|
47
|
-
npx --yes copilotkit@4.
|
|
47
|
+
npx --yes copilotkit@4.13.1 onboard read feature/stop
|
|
48
48
|
```
|
|
49
49
|
|
|
50
50
|
When the audit passes, run
|
|
51
|
-
`npx --yes copilotkit@4.
|
|
51
|
+
`npx --yes copilotkit@4.13.1 onboard read feature/complete` with the actual surface
|
|
52
52
|
outcome.
|