@plumpslabs/kuma 2.0.5 → 2.1.1
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 +41 -15
- package/dist/index.js +2346 -798
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -10,7 +10,9 @@
|
|
|
10
10
|
[](LICENSE)
|
|
11
11
|
[](https://nodejs.org/)
|
|
12
12
|
|
|
13
|
-
Works with **Claude Code
|
|
13
|
+
Works with **13 AI coding agents** — Claude Code, Cursor, Windsurf, GitHub Copilot Editor, Cline, Aider, Antigravity CLI, OpenCode, Codex CLI, Qwen Code, Kiro, OpenClaw, CodeWhale — and any MCP-compatible client.
|
|
14
|
+
|
|
15
|
+
**No manual config needed.** Just run `npx @plumpslabs/kuma init`.
|
|
14
16
|
|
|
15
17
|
</div>
|
|
16
18
|
|
|
@@ -18,7 +20,18 @@ Works with **Claude Code**, **Cursor**, **Gemini CLI**, **GitHub Copilot**, and
|
|
|
18
20
|
|
|
19
21
|
## Quick Start
|
|
20
22
|
|
|
21
|
-
|
|
23
|
+
```bash
|
|
24
|
+
# Generate config files for ALL supported AI agents
|
|
25
|
+
npx @plumpslabs/kuma init --all
|
|
26
|
+
|
|
27
|
+
# Or generate for specific agents
|
|
28
|
+
npx @plumpslabs/kuma init --cursor --claude --aider
|
|
29
|
+
|
|
30
|
+
# See all options
|
|
31
|
+
npx @plumpslabs/kuma init --help
|
|
32
|
+
```
|
|
33
|
+
|
|
34
|
+
Or add Kuma MCP server manually to any MCP client:
|
|
22
35
|
|
|
23
36
|
```json
|
|
24
37
|
{
|
|
@@ -31,18 +44,30 @@ Add to your MCP client config:
|
|
|
31
44
|
}
|
|
32
45
|
```
|
|
33
46
|
|
|
47
|
+
---
|
|
34
48
|
|
|
35
|
-
|
|
36
|
-
|
|
37
|
-
|
|
38
|
-
|
|
39
|
-
|
|
40
|
-
|
|
41
|
-
| **
|
|
42
|
-
|
|
|
43
|
-
|
|
|
44
|
-
|
|
45
|
-
|
|
49
|
+
## Supported Agents
|
|
50
|
+
|
|
51
|
+
`kuma init` generates native config files for **13 AI coding agents** — no manual hunting for file formats:
|
|
52
|
+
|
|
53
|
+
| # | Agent | Generated Files | Approach |
|
|
54
|
+
|---|-------|----------------|----------|
|
|
55
|
+
| 1 | **Claude Code** | `CLAUDE.md` | Fallback instructions (plugin via `/plugin install` is proper) |
|
|
56
|
+
| 2 | **Cursor** | `.cursor/rules/kuma.mdc` | Rule file with YAML frontmatter (`alwaysApply: true`) |
|
|
57
|
+
| 3 | **Windsurf** | `.windsurfrules` | Static rules file |
|
|
58
|
+
| 4 | **GitHub Copilot Editor** | `AGENTS.md` + `.github/skills/kuma/SKILL.md` | AGENTS.md + Skill file |
|
|
59
|
+
| 5 | **Cline** | `.clinerules/kuma.md` | Rule file with `paths` frontmatter |
|
|
60
|
+
| 6 | **Aider** | `CONVENTIONS.md` + `.aider.conf.yml` | Convention file referenced via `read: CONVENTIONS.md` |
|
|
61
|
+
| 7 | **Antigravity CLI** | `.agents/skills/kuma/SKILL.md` + `.agents/mcp_config.json` | Skill (loaded on demand) |
|
|
62
|
+
| 8 | **OpenCode** | `opencode.json` | Plugin config JSON |
|
|
63
|
+
| 9 | **Codex CLI (OpenAI)** | `AGENTS.md` + `.codex/config.toml` | AGENTS.md + MCP server in TOML |
|
|
64
|
+
| 10 | **Qwen Code** | `AGENTS.md` + `settings.json` | AGENTS.md + MCP server in JSON |
|
|
65
|
+
| 11 | **Kiro** | `.kiro/steering/kuma.md` | Steering file with YAML frontmatter (`inclusion: always`) |
|
|
66
|
+
| 12 | **OpenClaw** | `skills/kuma/SKILL.md` | Skill (loaded on demand) |
|
|
67
|
+
| 13 | **CodeWhale** | `skills/kuma/SKILL.md` + `.codewhale/mcp.json` | Skill + MCP server config |
|
|
68
|
+
|
|
69
|
+
> `AGENTS.md` includes sections for all selected agents that read it (Codex CLI, Qwen Code, GitHub Copilot Editor) — one file, no conflicts.
|
|
70
|
+
> `skills/kuma/SKILL.md` is shared by agents that load skills from workspace root.
|
|
46
71
|
|
|
47
72
|
---
|
|
48
73
|
|
|
@@ -64,7 +89,7 @@ Most tools make AI smarter. **Kuma makes AI not break things.**
|
|
|
64
89
|
|
|
65
90
|
---
|
|
66
91
|
|
|
67
|
-
## Tools (
|
|
92
|
+
## Tools (17)
|
|
68
93
|
|
|
69
94
|
### 🔍 Context — Understand the codebase
|
|
70
95
|
|
|
@@ -101,6 +126,7 @@ Most tools make AI smarter. **Kuma makes AI not break things.**
|
|
|
101
126
|
| `search_session_memory` | **Keyword search** across tool calls, memory files, errors, modified files, and dependency graph. |
|
|
102
127
|
| `write_memory` | Persist project knowledge (decisions, glossary) to `.kuma/memories/`. Append, prepend, or overwrite. |
|
|
103
128
|
| `kuma_reflect` | **Reflection tool** — checks if you're on track, detects drift (edits without tests, loops, unresolved failures), and suggests the next action. |
|
|
129
|
+
| `kuma_context` | **Snapshot manager** — save/restore project state (modified files, errors, git diff) before risky operations. |
|
|
104
130
|
|
|
105
131
|
---
|
|
106
132
|
|
|
@@ -136,7 +162,7 @@ Most tools make AI smarter. **Kuma makes AI not break things.**
|
|
|
136
162
|
2. **Safety first** — Every tool has a safety net: timeout, circuit breaker, rollback, sandbox.
|
|
137
163
|
3. **Graceful degradation, not crash** — Every tool has a fallback before it fails. LSP unavailable? Regex. File not found? Show resolved paths. Diff mismatch? Whitespace→fuzzy retry. Test fails? Circuit breaker stops the loop.
|
|
138
164
|
4. **Opinionated workflow** — Tools designed to be used together: `conventions → grep → pick → diff → test → review`.
|
|
139
|
-
5. **Minimal surface** —
|
|
165
|
+
5. **Minimal surface** — 17 focused tools. Each tool has one job and does it well. No overlap, no confusion.
|
|
140
166
|
|
|
141
167
|
---
|
|
142
168
|
|