substrate-ai 0.2.38 → 0.2.39
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.
|
@@ -8,12 +8,12 @@ This project uses Substrate for automated implementation pipelines. **When the u
|
|
|
8
8
|
**Just run it.** Substrate auto-detects which pipeline phase to start from (analysis → planning → solutioning → implementation) and auto-discovers pending stories. You do not need to determine the phase or find story keys manually.
|
|
9
9
|
|
|
10
10
|
```
|
|
11
|
-
substrate
|
|
11
|
+
substrate run --events
|
|
12
12
|
```
|
|
13
13
|
|
|
14
14
|
To target specific stories (if the user names them):
|
|
15
15
|
```
|
|
16
|
-
substrate
|
|
16
|
+
substrate run --events --stories 1-1,1-2,1-3
|
|
17
17
|
```
|
|
18
18
|
|
|
19
19
|
If substrate needs input it can't auto-detect (e.g., a project concept for analysis), it will exit with a clear error message telling you what to provide.
|
|
@@ -36,13 +36,18 @@ Check process health if pipeline seems quiet:
|
|
|
36
36
|
substrate health --output-format json
|
|
37
37
|
```
|
|
38
38
|
|
|
39
|
+
For long-running pipelines, attach the **supervisor** for automatic stall detection, kill-and-restart recovery, and post-run analysis. The supervisor monitors an active run — it does not start one. Start it alongside `substrate run`:
|
|
40
|
+
```
|
|
41
|
+
substrate supervisor --output-format json
|
|
42
|
+
```
|
|
43
|
+
|
|
39
44
|
**Interpreting silence:** No output for 5 minutes = normal (agent is working). No output for 15+ minutes = likely stalled. Use `substrate health` to confirm, then consider killing and resuming.
|
|
40
45
|
|
|
41
46
|
### After Pipeline Completes
|
|
42
47
|
|
|
43
48
|
1. **Summarize results** conversationally: X succeeded, Y failed, Z escalated
|
|
44
49
|
2. **Check metrics**: `substrate metrics --output-format json`
|
|
45
|
-
3. **Read analysis** (if supervisor
|
|
50
|
+
3. **Read analysis** (if supervisor was attached): `substrate metrics --analysis <run_id> --output-format json`
|
|
46
51
|
|
|
47
52
|
### Handling Escalations and Failures
|
|
48
53
|
|
|
@@ -57,7 +62,7 @@ substrate health --output-format json
|
|
|
57
62
|
| Command | Purpose |
|
|
58
63
|
|---|---|
|
|
59
64
|
| `substrate run --events` | Run pipeline with NDJSON event stream |
|
|
60
|
-
| `substrate supervisor --output-format json` |
|
|
65
|
+
| `substrate supervisor --output-format json` | Monitor active run with auto-recovery and post-run analysis |
|
|
61
66
|
| `substrate status --output-format json` | Poll current pipeline state |
|
|
62
67
|
| `substrate health --output-format json` | Check process health and stall detection |
|
|
63
68
|
| `substrate metrics --output-format json` | View historical run metrics |
|