@maestria/kimi-code 0.4.6 → 0.4.7
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 +13 -8
- package/package.json +1 -1
- package/skills/commands/blitz/SKILL.md +18 -0
- package/skills/commands/fein/SKILL.md +18 -0
- package/skills/commands/sonar/SKILL.md +18 -0
package/README.md
CHANGED
|
@@ -6,12 +6,12 @@ A declarative, manifest-based Kimi Code plugin that ships 8 specialized skills (
|
|
|
6
6
|
|
|
7
7
|
See [INSTALL.md](./INSTALL.md) for the full checklist. Quick start:
|
|
8
8
|
|
|
9
|
-
```
|
|
10
|
-
|
|
11
|
-
|
|
9
|
+
```bash
|
|
10
|
+
pnpx maestria@latest install kimi-code
|
|
11
|
+
pnpx maestria@latest status
|
|
12
12
|
```
|
|
13
13
|
|
|
14
|
-
|
|
14
|
+
The CLI pulls `@maestria/kimi-code` from npm and extracts it into `~/.kimi-code/plugins/managed/maestria`.
|
|
15
15
|
|
|
16
16
|
## The 8 Skills at a Glance
|
|
17
17
|
|
|
@@ -62,15 +62,21 @@ Unlike OpenCode's plugin SDK, Kimi Code's plugin system is **declarative** - no
|
|
|
62
62
|
- **No programmatic hooks** - the orchestrator skill carries the methodology, and Kimi Code's `[[hooks]]` blocks (user-managed) cover the rest.
|
|
63
63
|
- **No custom subagent identity** - Kimi Code hardcodes `coder`/`explore`/`plan`. The 7 specialist identities are encoded as persona content in prompt templates.
|
|
64
64
|
|
|
65
|
-
See [
|
|
65
|
+
See the [architecture documentation](https://maestria.sznm.dev/kimi-code/architecture/) for the full design rationale.
|
|
66
66
|
|
|
67
67
|
## Updating
|
|
68
68
|
|
|
69
|
+
```bash
|
|
70
|
+
pnpx maestria@latest update kimi-code
|
|
69
71
|
```
|
|
70
|
-
|
|
72
|
+
|
|
73
|
+
Updates follow the latest release by default. Pin to a specific version with:
|
|
74
|
+
|
|
75
|
+
```bash
|
|
76
|
+
pnpx maestria@latest update kimi-code@0.4.6
|
|
71
77
|
```
|
|
72
78
|
|
|
73
|
-
|
|
79
|
+
See [INSTALL.md](./INSTALL.md) for details.
|
|
74
80
|
|
|
75
81
|
## Contributing
|
|
76
82
|
|
|
@@ -83,4 +89,3 @@ MIT
|
|
|
83
89
|
## Related
|
|
84
90
|
|
|
85
91
|
- [`@maestria/opencode`](../opencode/README.md) - the OpenCode variant of this plugin (TypeScript SDK, programmatic hooks).
|
|
86
|
-
- [ADR-KC-001](../../docs/adr/kimi-code/ADR-KC-001-kimi-code-architecture.md) - the architecture decision record for the Kimi Code plugin.
|
package/package.json
CHANGED
|
@@ -0,0 +1,18 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: blitz
|
|
3
|
+
description: "Fast implementation mode: skip recon/design unless unknown"
|
|
4
|
+
type: prompt
|
|
5
|
+
whenToUse: When the user types /blitz or includes "blitz" in their message for fast implementation.
|
|
6
|
+
arguments: []
|
|
7
|
+
---
|
|
8
|
+
|
|
9
|
+
<!-- Auto-generated from @maestria/core. Do not edit directly.
|
|
10
|
+
Edit the canonical file at packages/core/agent-directives/ instead. -->
|
|
11
|
+
|
|
12
|
+
**Skill profile:** `plan` - workflow mode command. You have Read, Glob, Grep, Bash, FetchURL, and WebSearch.
|
|
13
|
+
|
|
14
|
+
[MODE: blitz]
|
|
15
|
+
|
|
16
|
+
## MODE: blitz (Fast Implementation)
|
|
17
|
+
|
|
18
|
+
Speed mode: skip reconnaissance and design gates. Go directly to builder for implementation. Only use adventurer if the codebase context is genuinely unknown (not as a default step). Skip reviewer unless the user explicitly requests review.
|
|
@@ -0,0 +1,18 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: fein
|
|
3
|
+
description: "Full pipeline mode: recon, design, implement, review"
|
|
4
|
+
type: prompt
|
|
5
|
+
whenToUse: When the user types /fein or includes "fein" in their message to run the complete maestria pipeline.
|
|
6
|
+
arguments: []
|
|
7
|
+
---
|
|
8
|
+
|
|
9
|
+
<!-- Auto-generated from @maestria/core. Do not edit directly.
|
|
10
|
+
Edit the canonical file at packages/core/agent-directives/ instead. -->
|
|
11
|
+
|
|
12
|
+
**Skill profile:** `plan` - workflow mode command. You have Read, Glob, Grep, Bash, FetchURL, and WebSearch.
|
|
13
|
+
|
|
14
|
+
[MODE: fein]
|
|
15
|
+
|
|
16
|
+
## MODE: fein (Full Pipeline)
|
|
17
|
+
|
|
18
|
+
Default role-based pipeline: thinker (recon/design/plan) -> worker (implementation) -> verifier (review). Verifier acceptance terminates the pipeline for that unit of work. Roles and order may adapt to task needs - this is the default, not a fixed requirement. Do NOT skip any phase unless the user explicitly overrides in the same turn.
|
|
@@ -0,0 +1,18 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: sonar
|
|
3
|
+
description: "Research-only mode: recon and design, no implementation"
|
|
4
|
+
type: prompt
|
|
5
|
+
whenToUse: When the user types /sonar or includes "sonar" in their message for research-only work.
|
|
6
|
+
arguments: []
|
|
7
|
+
---
|
|
8
|
+
|
|
9
|
+
<!-- Auto-generated from @maestria/core. Do not edit directly.
|
|
10
|
+
Edit the canonical file at packages/core/agent-directives/ instead. -->
|
|
11
|
+
|
|
12
|
+
**Skill profile:** `plan` - workflow mode command. You have Read, Glob, Grep, Bash, FetchURL, and WebSearch.
|
|
13
|
+
|
|
14
|
+
[MODE: sonar]
|
|
15
|
+
|
|
16
|
+
## MODE: sonar (Research Only)
|
|
17
|
+
|
|
18
|
+
Research mode: reconnaissance and design only. Delegate to adventurer (recon) followed by architect or planner (analysis/design). STOP after delivering findings and design. Do NOT implement, write code, or create any production files.
|