karukia-mcp 1.2.0

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/LICENSE ADDED
@@ -0,0 +1,37 @@
1
+ KARUKIA MCP - License
2
+
3
+ Copyright (c) 2026 KARUKIA (contact@karukia.com)
4
+
5
+ Permission is hereby granted, free of charge, to any person or organization
6
+ ("User") to use this software and associated knowledge base ("KARUKIA MCP")
7
+ for personal, educational, and internal professional purposes, subject to
8
+ the following conditions:
9
+
10
+ 1. USAGE
11
+ KARUKIA MCP runs locally on the User's machine via the MCP protocol.
12
+ No account or registration is required.
13
+
14
+ 2. COMMERCIAL USE
15
+ Any commercial use, resale, redistribution for profit, or integration of
16
+ KARUKIA MCP (in whole or in part) into a paid product or service requires
17
+ prior written authorization from KARUKIA.
18
+
19
+ Contact: contact@karukia.com
20
+
21
+ 3. ATTRIBUTION
22
+ All copies or substantial portions of this software must retain this
23
+ license notice and the KARUKIA copyright.
24
+
25
+ 4. MODIFICATION
26
+ Users may fork and modify this software for their own internal use.
27
+ Public redistribution of modified versions requires written authorization
28
+ from KARUKIA (contact@karukia.com).
29
+
30
+ 5. NO WARRANTY
31
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
32
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
33
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
34
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
35
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
36
+ FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
37
+ DEALINGS IN THE SOFTWARE.
package/README.md ADDED
@@ -0,0 +1,232 @@
1
+ # KARUKIA MCP v1.2
2
+
3
+ **The complete AI-assisted development methodology, delivered via MCP.**
4
+
5
+ 21 tools, 11 skills, 935+ security/quality/pentest checkpoints. Works with any AI platform (Claude Code, Cursor, Windsurf, Copilot...) through the Model Context Protocol.
6
+
7
+ ## What is KARUKIA?
8
+
9
+ KARUKIA is a structured development methodology built around specialized AI personas. Each persona (Neo for security, Jeffrey for architecture, Viper for pentesting, Opo for quality...) comes with its own workflow, guard rails, and knowledge base.
10
+
11
+ When you call a KARUKIA tool, the MCP server returns a complete prompt — persona identity, workflow, checklists, templates — that transforms your AI assistant into that specialist for the session.
12
+
13
+ ```
14
+ You: "Run a security audit"
15
+ -> AI calls neo tool
16
+ -> MCP returns full Neo persona prompt + 445 security controls inline
17
+ -> AI becomes Neo, follows the methodology, produces structured findings
18
+ ```
19
+
20
+ ## The Three Layers
21
+
22
+ ```
23
+ Layer 1 - DEFENSIVE (Neo) 445 controls "Is my code secure?"
24
+ Layer 2 - QUALITY (Opquast) 245 rules "Is my app well-built?"
25
+ Layer 3 - OFFENSIVE (Viper) 245 tests "How would a hacker break in?"
26
+ ```
27
+
28
+ ---
29
+
30
+ ## Quick Start
31
+
32
+ **Prerequisites:** [Node.js](https://nodejs.org/) 20 or later.
33
+
34
+ ### Step 1 — Add KARUKIA to your project
35
+
36
+ Create or edit `.mcp.json` at the root of your project:
37
+
38
+ ```json
39
+ {
40
+ "mcpServers": {
41
+ "karukia": {
42
+ "command": "npx",
43
+ "args": ["karukia-mcp"]
44
+ }
45
+ }
46
+ }
47
+ ```
48
+
49
+ > **Note:** If the file already exists and has other MCP servers, just add the `"karukia"` key inside the existing `"mcpServers"` object.
50
+
51
+ ### Step 2 — Restart your AI client
52
+
53
+ Restart Claude Code (`/quit` then relaunch) or your IDE. The 21 KARUKIA tools are now available.
54
+
55
+ > On first launch, `npx` downloads the package automatically (~175 KB). Subsequent launches use the cached version.
56
+
57
+ ### Step 3 — Configure your project
58
+
59
+ Tell your AI:
60
+
61
+ > "Call the `install` tool from karukia"
62
+
63
+ KARUKIA scans your project, detects your stack, and generates configuration files (security scope, CLAUDE.md, memory structure).
64
+
65
+ ### Step 4 — Start working
66
+
67
+ > "Use /auto to add user authentication"
68
+ > "Use /auto to audit security"
69
+ > "Use /auto to run a pentest"
70
+
71
+ `/auto` is your daily driver. It routes your request to the right specialists automatically.
72
+
73
+ > **Tip:** You only need two tools: `/install` (once) then `/auto` (always). For direct control, call `/neo`, `/viper`, or `/jeffrey` individually. Call `start` anytime for a full guide.
74
+
75
+ ### Where to put the config
76
+
77
+ | Client | File | Scope |
78
+ |--------|------|-------|
79
+ | **Claude Code CLI** | `.mcp.json` at project root | This project only |
80
+ | **Claude Code CLI** | `~/.claude.json` (home directory) | All your projects |
81
+ | **Claude Desktop** | `claude_desktop_config.json` | Global |
82
+ | **Cursor** | `.cursor/mcp.json` at project root | This project only |
83
+ | **Windsurf** | MCP settings panel | Global |
84
+
85
+ ---
86
+
87
+ ## Global Installation (optional)
88
+
89
+ If you want KARUKIA available in all your projects without adding `.mcp.json` each time:
90
+
91
+ ```bash
92
+ npm install -g karukia-mcp
93
+ ```
94
+
95
+ Then add to your global AI config (`~/.claude.json` for Claude Code):
96
+
97
+ ```json
98
+ {
99
+ "mcpServers": {
100
+ "karukia": {
101
+ "command": "karukia-mcp"
102
+ }
103
+ }
104
+ }
105
+ ```
106
+
107
+ ---
108
+
109
+ ## 21 Tools
110
+
111
+ ### Essential (start here)
112
+
113
+ | Tool | Description |
114
+ |------|-------------|
115
+ | `install` | **[FIRST STEP]** Configure KARUKIA for your project — run once |
116
+ | `auto` | **[MAIN TOOL]** Describe what you need — KARUKIA routes to the right skills |
117
+ | `start` | Quick-start guide — explains all skills at 3 progressive levels |
118
+
119
+ ### 11 Skills (AI Personas)
120
+
121
+ Each skill returns a complete prompt that transforms your AI into a specialist.
122
+
123
+ | Tool | Persona | What it does |
124
+ |------|---------|-------------|
125
+ | `neo` | Security Auditor | Defensive audit against 6 frameworks (OWASP, HDS, ISO 27001, SOC 2, PCI-DSS, HIPAA) |
126
+ | `viper` | Pentest Brigade | Offensive testing with 16 agents, CVSS v4 scoring, MITRE ATT&CK mapping |
127
+ | `jeffrey` | Full-Stack Architect | Feature implementation with TDD and security validation |
128
+ | `opo` | Quality Validator | Web quality against 245 Opquast rules |
129
+ | `audit_opquast` | Quality Auditor | Deep Opquast compliance audit with 14 thematic checklists |
130
+ | `ebios_rm_audit` | Risk Analyst | EBIOS Risk Manager methodology (ANSSI) — formal risk analysis |
131
+ | `security_hardening` | Hardening Planner | Security improvement chantiers |
132
+ | `terraform_update` | IaC Specialist | Terraform automation for KMS, GCS, IAM |
133
+ | `doc_refactor` | Doc Auditor | Documentation accuracy audit vs actual code |
134
+
135
+ ### 5 Utilities
136
+
137
+ | Tool | Description |
138
+ |------|-------------|
139
+ | `list_checklists` | Browse all 24 checklists by category |
140
+ | `get_checklist` | Retrieve the full content of any checklist |
141
+ | `search_rules` | Search across all 935+ checkpoints by keyword and severity |
142
+ | `suggest_checklists` | Describe your project — get a prioritized audit plan |
143
+ | `generate_report` | Compile audit results into a scored Markdown report |
144
+
145
+ ### 4 Memory & Config
146
+
147
+ | Tool | Description |
148
+ |------|-------------|
149
+ | `init_memory` | Initialize KARUKIA memory structure in a project |
150
+ | `get_session_template` | Get pre-filled session templates for any skill |
151
+ | `get_config_template` | Get configuration templates (security scope, CLAUDE.md, analytics) |
152
+ | `get_shared` | Access shared methodology components (guard rules, workflow, agents) |
153
+
154
+ ---
155
+
156
+ ## 24 Checklists
157
+
158
+ ### Defensive Security (Neo) — 6 checklists, 445 controls
159
+
160
+ | Checklist | Points | Scope |
161
+ |-----------|--------|-------|
162
+ | **OWASP Security Baseline** | 62 | Every web app |
163
+ | **HDS 2.0** | 52 | Health data, France |
164
+ | **ISO 27001:2022** | 93 | Enterprise ISMS |
165
+ | **SOC 2 Type II** | 74 | SaaS, US market |
166
+ | **PCI-DSS v4.0** | 97 | Payment processing |
167
+ | **HIPAA** | 67 | Health data, US |
168
+
169
+ ### Web Quality (Opquast) — 14 checklists, 245 rules
170
+
171
+ Content, personal data, e-commerce, forms, identity, images, internationalization, links, navigation, newsletter, presentation, security UX, server performance, and code structure.
172
+
173
+ Based on [Opquast](https://www.opquast.com/) — the French web quality reference used by 15,000+ professionals.
174
+
175
+ ### Offensive Security (Viper) — 4 checklists, 245+ tests
176
+
177
+ | Checklist | Tests | Scope |
178
+ |-----------|-------|-------|
179
+ | **OWASP WSTG v5** | 100 | Web penetration testing |
180
+ | **Cloud Platform** | 80+ | Firebase, GCP, AWS, Azure |
181
+ | **Healthcare** | 50+ | PHI, encryption, medical data |
182
+ | **Attack Scenarios** | 15+ | PTES templates, MITRE ATT&CK |
183
+
184
+ ---
185
+
186
+ ## Usage Examples
187
+
188
+ ### Full security audit
189
+
190
+ > "Run a security audit on my project"
191
+
192
+ Your AI calls `neo` — becomes the Neo security auditor — follows the methodology — produces structured findings with severity, file:line references, and remediation steps.
193
+
194
+ ### Build a feature with guardrails
195
+
196
+ > "Implement user authentication with /jeffrey"
197
+
198
+ Your AI calls `jeffrey` — becomes the Jeffrey architect — implements with TDD, then chains to Neo for security validation (rejection loop: if Neo rejects, Jeffrey fixes, max 3 iterations).
199
+
200
+ ### Pentest your app
201
+
202
+ > "Run a pentest with /viper"
203
+
204
+ Your AI calls `viper` — deploys the Brigade methodology with 16 specialized agents across Recon, Surface Analysis, and Exploitation phases.
205
+
206
+ ### Orchestrate everything
207
+
208
+ > "Use /auto to handle my request"
209
+
210
+ Your AI calls `auto` — analyzes the request — routes to the right skill(s) — manages the chain.
211
+
212
+ ---
213
+
214
+ ## About KARUKIA
215
+
216
+ KARUKIA is a structured AI-assisted development methodology built around three principles:
217
+
218
+ 1. **Separation of concerns** — Security, quality, and implementation are separate disciplines handled by separate AI personas.
219
+ 2. **Formal checkpoints over gut feeling** — 935 documented checkpoints beat "I think it's fine."
220
+ 3. **Defense in depth** — Defensive audit first, quality validation second, offensive testing last.
221
+
222
+ Built from real-world experience securing a healthcare SaaS application to HDS 2.0 / ISO 27001 standards.
223
+
224
+ ---
225
+
226
+ ## License
227
+
228
+ KARUKIA MCP is free for personal, educational, and internal professional use.
229
+
230
+ **Commercial use or resale requires written authorization.** Contact: contact@karukia.com
231
+
232
+ See [LICENSE](./LICENSE) for full terms.