@kimbho/kimbho-cli 0.1.3 → 0.1.5
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 +24 -4
- package/dist/index.cjs +2311 -1227
- package/dist/index.cjs.map +4 -4
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -36,12 +36,13 @@ kimbho plan "build a coding agent"
|
|
|
36
36
|
kimbho /models openrouter --search claude --limit 10
|
|
37
37
|
```
|
|
38
38
|
|
|
39
|
-
Bare `kimbho` opens the interactive shell. Within that shell, plain-English input is
|
|
39
|
+
Bare `kimbho` opens the interactive shell. Within that shell, plain-English input is sent to the active model, and slash commands like `/plan`, `/run`, `/models`, and `/brains` are supported. Provider and model selections apply to all agent roles by default.
|
|
40
40
|
|
|
41
41
|
Useful shell commands:
|
|
42
42
|
|
|
43
43
|
```bash
|
|
44
44
|
/status
|
|
45
|
+
/ask say hello from lm studio
|
|
45
46
|
/providers
|
|
46
47
|
/providers add openrouter
|
|
47
48
|
/providers use openrouter-main
|
|
@@ -51,14 +52,33 @@ Useful shell commands:
|
|
|
51
52
|
/model
|
|
52
53
|
/plan build a coding agent
|
|
53
54
|
/run scaffold a SaaS starter
|
|
54
|
-
/resume --execute --max-auto-tasks 2
|
|
55
|
+
/resume --execute --max-auto-tasks 2 --max-repair-attempts 2
|
|
55
56
|
/quit
|
|
56
57
|
```
|
|
57
58
|
|
|
58
|
-
|
|
59
|
+
LM Studio example:
|
|
60
|
+
|
|
61
|
+
```bash
|
|
62
|
+
/providers add lmstudio lmstudio-lan --base-url http://192.168.0.70:1234
|
|
63
|
+
/models
|
|
64
|
+
/select 1
|
|
65
|
+
```
|
|
66
|
+
|
|
67
|
+
`run` now auto-executes the supported ready-task frontier and persists session events. Specialist tasks are executed through a bounded model-driven tool loop that can list files, search source contents, read files, write files, apply patches, inspect git diff, run verification commands, and emit resumable transcripts. `resume --execute` continues that frontier from the latest session snapshot.
|
|
68
|
+
After the agent changes workspace state, it now has to pass a verification step before the task can finish.
|
|
69
|
+
If verification fails, Kimbho forces a repair pass before the next rerun and blocks the task after the configured repair budget is exhausted.
|
|
70
|
+
If a non-debugger specialist burns through that budget, Kimbho automatically hands the task to `test-debugger` and keeps the same task id/dependency chain intact.
|
|
71
|
+
|
|
72
|
+
Useful tuning flags:
|
|
73
|
+
|
|
74
|
+
```bash
|
|
75
|
+
kimbho run "fix the failing test suite" --max-agent-steps 10 --max-repair-attempts 3
|
|
76
|
+
kimbho resume --execute --max-auto-tasks 2 --max-repair-attempts 3
|
|
77
|
+
```
|
|
59
78
|
|
|
60
79
|
## Notes
|
|
61
80
|
|
|
62
81
|
- The published CLI bundles the internal Kimbho packages so installation does not depend on the monorepo layout.
|
|
63
82
|
- Brain and provider configuration is stored in `.kimbho/config.json`.
|
|
64
|
-
- Use `kimbho models use <model> --provider <id> --
|
|
83
|
+
- Use `kimbho models use <model> --provider <id> --set-default` to select a provider model for the whole system.
|
|
84
|
+
- Use `kimbho brains assign --role <role> ...` only when you intentionally want an advanced per-role override.
|