@deimoscloud/coreai 0.1.0 → 0.1.2

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 CHANGED
@@ -19,17 +19,26 @@ npm install -g @deimoscloud/coreai
19
19
  ## Quick Start
20
20
 
21
21
  ```bash
22
- # Initialize a new project (coming soon)
22
+ # Initialize a new project
23
23
  coreai init
24
24
 
25
25
  # Build agents to .claude/agents/
26
26
  coreai build
27
27
 
28
+ # Build agents and initialize KnowledgeLibrary
29
+ coreai build --init-knowledge-library
30
+
31
+ # Generate Claude skills (slash commands)
32
+ coreai skills generate
33
+
28
34
  # List available agents
29
35
  coreai agents list
30
36
 
31
37
  # Show details for a specific agent
32
38
  coreai agents show backend-engineer
39
+
40
+ # Check agent status
41
+ coreai status
33
42
  ```
34
43
 
35
44
  ## Commands
@@ -79,6 +88,42 @@ coreai agents show backend-engineer --json
79
88
  coreai agents show backend-engineer --markdown
80
89
  ```
81
90
 
91
+ ### `coreai status`
92
+
93
+ Show agent states and pending messages.
94
+
95
+ ```bash
96
+ # Show all agent states
97
+ coreai status
98
+
99
+ # Show detailed message information
100
+ coreai status --detailed
101
+
102
+ # Show status for a specific agent
103
+ coreai status --agent backend-engineer
104
+
105
+ # Initialize KnowledgeLibrary if not exists
106
+ coreai status --init
107
+ ```
108
+
109
+ ### `coreai skills`
110
+
111
+ Manage Claude skills (slash commands).
112
+
113
+ ```bash
114
+ # Generate skills to .claude/commands/
115
+ coreai skills generate
116
+
117
+ # List available skills
118
+ coreai skills list
119
+
120
+ # Generate specific skills only
121
+ coreai skills generate --skills implement-feature,create-pr
122
+
123
+ # Overwrite existing skill files
124
+ coreai skills generate --overwrite
125
+ ```
126
+
82
127
  ### Other Commands
83
128
 
84
129
  ```bash
@@ -88,16 +133,14 @@ coreai --version
88
133
  # Show help
89
134
  coreai --help
90
135
 
91
- # Validate configuration (coming soon)
136
+ # Validate configuration
92
137
  coreai validate
93
138
 
94
- # Sync remote context (coming soon)
139
+ # Sync remote context from configured sources
95
140
  coreai sync
141
+ coreai sync --force # Force refresh all cached content
96
142
 
97
- # Show project status (coming soon)
98
- coreai status
99
-
100
- # Cache management (coming soon)
143
+ # Cache management
101
144
  coreai cache status
102
145
  coreai cache clear
103
146
  ```
@@ -210,27 +253,184 @@ description: "Backend engineer for ${config.project.name}"
210
253
  description: "I am ${agent.display_name}, responsible for ${agent.role}"
211
254
  ```
212
255
 
213
- ### `${remote.*}` - Remote Content (Coming Soon)
256
+ ### `${remote.*}` - Remote Content
214
257
 
215
258
  ```yaml
216
259
  context:
217
260
  - "${remote.confluence.architecture-docs}"
218
261
  ```
219
262
 
263
+ ## Skills (Claude Slash Commands)
264
+
265
+ CoreAI generates Claude Code slash commands that integrate with your project configuration.
266
+
267
+ ### Built-in Skills
268
+
269
+ | Skill | Description |
270
+ |-------|-------------|
271
+ | `implement-feature` | Implement a feature following project workflow |
272
+ | `create-pr` | Create a pull request with proper formatting |
273
+ | `review-code` | Review code changes for quality and standards |
274
+ | `run-quality-gates` | Run configured lint, test, and build commands |
275
+
276
+ ### Generating Skills
277
+
278
+ ```bash
279
+ # Generate all skills
280
+ coreai skills generate
281
+
282
+ # List available skills
283
+ coreai skills list
284
+
285
+ # Generate to custom directory
286
+ coreai skills generate --output ./my-commands
287
+ ```
288
+
289
+ Generated skills are placed in `.claude/commands/` and become available as `/skill-name` in Claude Code.
290
+
291
+ ### Custom Skills
292
+
293
+ Create custom skills in `coreai/skills/` directory:
294
+
295
+ ```markdown
296
+ ---
297
+ name: deploy
298
+ description: Deploy the application to staging
299
+ category: custom
300
+ ---
301
+
302
+ Deploy the application following the deployment checklist:
303
+
304
+ 1. Run all quality gates
305
+ 2. Build the production bundle
306
+ 3. Deploy to staging environment
307
+ 4. Run smoke tests
308
+ ```
309
+
310
+ ## MCP Server Integration
311
+
312
+ CoreAI integrates with Model Context Protocol (MCP) servers for enhanced functionality.
313
+
314
+ ### Automatic Discovery
315
+
316
+ The CLI automatically discovers MCP servers from:
317
+
318
+ 1. **Project-level**: `mcp.json`, `.mcp.json`, `.claude/mcp.json`
319
+ 2. **Global config**: `~/.config/claude/mcp.json`, `~/.claude/mcp.json`
320
+ 3. **Claude Desktop**: `~/Library/Application Support/Claude/claude_desktop_config.json`
321
+
322
+ ### Supported MCP Servers
323
+
324
+ | Server | Integration |
325
+ |--------|-------------|
326
+ | `@modelcontextprotocol/server-github` | GitHub issues, PRs, repositories |
327
+ | `@anthropic/mcp-server-jira` | Jira issue tracking |
328
+ | `@anthropic/mcp-server-confluence` | Confluence documentation |
329
+ | `@modelcontextprotocol/server-filesystem` | Local file operations |
330
+
331
+ ### Configuration Example
332
+
333
+ Create `.claude/mcp.json` in your project:
334
+
335
+ ```json
336
+ {
337
+ "mcpServers": {
338
+ "github": {
339
+ "command": "npx",
340
+ "args": ["-y", "@modelcontextprotocol/server-github"],
341
+ "env": {
342
+ "GITHUB_TOKEN": "${GITHUB_TOKEN}"
343
+ }
344
+ },
345
+ "filesystem": {
346
+ "command": "npx",
347
+ "args": ["-y", "@modelcontextprotocol/server-filesystem", "./"]
348
+ }
349
+ }
350
+ }
351
+ ```
352
+
353
+ ### Native Fallbacks
354
+
355
+ When MCP servers are unavailable, CoreAI falls back to native adapters:
356
+
357
+ - **GitHub**: Uses `gh` CLI or GitHub REST API
358
+ - **Filesystem**: Direct file system operations
359
+
360
+ ## KnowledgeLibrary
361
+
362
+ The KnowledgeLibrary provides persistent state management for agents.
363
+
364
+ ### Structure
365
+
366
+ ```
367
+ KnowledgeLibrary/
368
+ ├── context.txt # Project context
369
+ ├── architecture.txt # Architecture overview
370
+ ├── prd.txt # Product requirements
371
+ └── backend-engineer/ # Per-agent directories
372
+ ├── inbox/ # Incoming messages
373
+ │ └── processed/ # Processed messages
374
+ ├── outbox/ # Outgoing messages
375
+ ├── context/ # Agent-specific context
376
+ │ └── current.txt # Current working state
377
+ ├── control/ # Control files
378
+ │ ├── objectives.txt # Current objectives
379
+ │ ├── decisions.txt # Key decisions
380
+ │ └── dependencies.txt # Dependencies
381
+ ├── history/ # Historical records
382
+ └── tech/ # Technical notes
383
+ ```
384
+
385
+ ### Initializing
386
+
387
+ ```bash
388
+ # Initialize during build
389
+ coreai build --init-knowledge-library
390
+
391
+ # Or initialize separately
392
+ coreai status --init
393
+ ```
394
+
395
+ ### Checking Status
396
+
397
+ ```bash
398
+ # View all agents
399
+ coreai status
400
+
401
+ # Detailed view with messages
402
+ coreai status --detailed
403
+
404
+ # JSON output for scripting
405
+ coreai status --json
406
+ ```
407
+
220
408
  ## Project Structure
221
409
 
222
410
  ```
223
411
  your-project/
224
412
  ├── coreai.config.yaml # Project configuration
225
413
  ├── coreai/
226
- └── agents/ # Custom agent definitions
227
- └── my-agent.yaml
414
+ ├── agents/ # Custom agent definitions
415
+ └── my-agent.yaml
416
+ │ └── skills/ # Custom skill templates
417
+ │ └── my-skill.md
228
418
  ├── .claude/
229
- └── agents/ # Compiled agents (generated)
230
- ├── backend-engineer.md
231
- └── my-agent.md
232
- └── .coreai/
233
- └── cache/ # Local cache (coming soon)
419
+ ├── mcp.json # MCP server configuration
420
+ ├── agents/ # Compiled agents (generated)
421
+ │ ├── backend-engineer.md
422
+ │ │ └── my-agent.md
423
+ └── commands/ # Generated skills (slash commands)
424
+ │ ├── implement-feature.md
425
+ │ └── create-pr.md
426
+ ├── .coreai/
427
+ │ └── cache/ # Remote content cache
428
+ └── KnowledgeLibrary/ # Agent state management
429
+ ├── context.txt
430
+ └── backend-engineer/
431
+ ├── inbox/
432
+ ├── outbox/
433
+ └── context/
234
434
  ```
235
435
 
236
436
  ## Development
@@ -244,7 +444,7 @@ your-project/
244
444
 
245
445
  ```bash
246
446
  # Clone the repository
247
- git clone https://github.com/your-org/coreai.git
447
+ git clone https://github.com/DeimosCloud/coreai.git
248
448
  cd coreai
249
449
 
250
450
  # Install dependencies
@@ -278,25 +478,27 @@ npm run test:watch
278
478
 
279
479
  See [IMPLEMENTATION_PLAN.md](IMPLEMENTATION_PLAN.md) for detailed implementation status.
280
480
 
281
- ### Completed
481
+ ### Phase 1 (Complete)
282
482
 
283
483
  - Project foundation (TypeScript, CLI, configuration)
284
484
  - Agent YAML schema and validation
285
485
  - Variable resolution system
286
486
  - Agent compiler (YAML to Markdown)
287
487
  - Agent CLI commands (build, list, show)
288
-
289
- ### In Progress
290
-
291
488
  - Integration adapter framework
292
- - MCP integration layer
489
+ - MCP integration layer with server discovery
490
+ - Shared context and caching
491
+ - Skill generation (Claude slash commands)
492
+ - KnowledgeLibrary (local agent state management)
493
+ - Complete CLI binary
293
494
 
294
- ### Planned
495
+ ### Phase 2 (Planned)
295
496
 
296
- - Shared context and caching
297
- - Remote content fetching
298
- - Skill generation
299
- - Local state management
497
+ - Real-time agent collaboration
498
+ - Advanced MCP server integrations
499
+ - Remote context syncing with Confluence/Notion
500
+ - Agent workflow orchestration
501
+ - Multi-project support
300
502
 
301
503
  ## License
302
504