@deimoscloud/coreai 0.1.19 → 0.1.20

package/README.md

@@ -4,8 +4,8 @@ A configurable, team-ready AI agent orchestration platform for Claude Code.
 
 ## Features
 
-- **YAML-based agent definitions** - Define agents in structured YAML, compile to Claude-compatible markdown
-- **Variable resolution** - Use `${config.*}`, `${agent.*}`, and `${remote.*}` variables in definitions
+- **Markdown-based agent definitions** - Define agents in rich Markdown templates with YAML frontmatter
+- **Variable resolution** - Use `${config.*}`, `${agent.*}`, and `${remote.*}` variables in templates
 - **Core + custom agents** - Built-in agent library with support for project-specific customizations
 - **Team filtering** - Configure which agents are available per project
 - **Override system** - Customize core agents without modifying the source
@@ -197,32 +197,38 @@ Create a `coreai.config.yaml` in your project root:
 ```yaml
 project:
   name: "My Project"
-  description: "Project description"
+  description: "Brief description of your project"
 
+# Default: [engineering-manager]
+# Add agents with: coreai agents add <name>
 team:
   agents:
+    - engineering-manager
     - backend-engineer
     - frontend-engineer
-    - devops-engineer
 
 integrations:
-  github:
-    repo: "owner/repo"
-  jira:
-    project: "PROJ"
+  git:
+    provider: github
+    config:
+      owner: "your-org"
+      repo: "your-repo"
+      default_branch: main
+  issue_tracker:
+    provider: jira
+    config:
+      project_key: "PROJ"
+      base_url: "https://your-org.atlassian.net"
 
 quality_gates:
-  lint: "npm run lint"
-  test: "npm test"
-  build: "npm run build"
-
-tech_stack:
-  languages:
-    - TypeScript
-    - Node.js
-  frameworks:
-    - Express
-    - React
+  lint:
+    command: npm run lint
+    required: true
+  test:
+    command: npm test
+    required: true
+  build:
+    command: npm run build
 ```
 
 ## Built-in Agents
@@ -249,71 +255,70 @@ CoreAI includes these core agents:
 
 ## Custom Agents
 
-Create custom agents in `coreai/agents/` directory:
+Create custom agents as Markdown files in `coreai/agents/` directory:
 
-```yaml
-# coreai/agents/mobile-engineer.yaml
-role: mobile-engineer
-type: ic-engineer
-display_name: "Mobile Engineer"
-description: "React Native mobile development specialist"
-
-responsibilities:
-  - Implement mobile features
-  - Ensure cross-platform compatibility
-  - Optimize mobile performance
-
-expertise:
-  primary:
-    - React Native
-    - iOS development
-    - Android development
-
-behaviors:
-  workflow: implementation
-  quality_gates:
-    - lint
-    - test
-    - build
+```markdown
+---
+name: mobile-engineer
+description: React Native mobile development specialist
+tools: Read, Write, Edit, Bash, Glob, Grep
+---
+
+# Mobile Engineer
+
+## Role
+
+React Native mobile development specialist focused on cross-platform mobile apps.
+
+---
+
+## Your Responsibilities
+
+- Implement mobile features using React Native
+- Ensure cross-platform compatibility (iOS and Android)
+- Optimize mobile performance and user experience
 ```
 
 ### Overriding Core Agents
 
 To customize a core agent, create a file with the same name in `coreai/agents/`:
 
-```yaml
-# coreai/agents/backend-engineer.yaml
-# This overrides the core backend-engineer agent
+```markdown
+---
+name: backend-engineer
+description: Custom backend engineer for our Python stack
+tools: Read, Write, Edit, Bash, Glob, Grep
+---
 
-role: backend-engineer
-type: ic-engineer
-display_name: "Backend Engineer"
-description: "Custom backend engineer for our Python stack"
+# Backend Engineer
 
-# ... rest of definition
+## Role
+
+Custom backend engineer specializing in our Python/FastAPI stack.
+
+<!-- Your custom agent content here -->
 ```
 
 ## Variable Resolution
 
-Agent definitions support three variable namespaces:
+Agent templates support three variable namespaces:
 
 ### `${config.*}` - Project Configuration
 
-```yaml
-description: "Backend engineer for ${config.project.name}"
+```markdown
+Working on project: ${config.project.name}
 ```
 
 ### `${agent.*}` - Agent Self-Reference
 
-```yaml
-description: "I am ${agent.display_name}, responsible for ${agent.role}"
+```markdown
+You are the ${agent.name}, responsible for ${agent.role}
 ```
 
 ### `${remote.*}` - Remote Content
 
-```yaml
-context:
-  - "${remote.confluence.architecture-docs}"
+```markdown
+Reference the architecture documentation: ${remote.documentation}
 ```
 
 ## Skills (Claude Slash Commands)
@@ -468,7 +473,7 @@ your-project/
 ├── coreai.config.yaml          # Project configuration
 ├── coreai/
 │   ├── agents/                 # Custom agent definitions
-│   │   └── my-agent.yaml
+│   │   └── my-agent.md
 │   └── skills/                 # Custom skill templates
 │       └── my-skill.md
 ├── .claude/
@@ -537,9 +542,9 @@ See [IMPLEMENTATION_PLAN.md](IMPLEMENTATION_PLAN.md) for detailed implementation
 ### Phase 1 (Complete)
 
 - Project foundation (TypeScript, CLI, configuration)
-- Agent YAML schema and validation
+- Agent schema and validation
 - Variable resolution system
-- Agent compiler (YAML to Markdown)
+- Agent compiler (Markdown templates to Claude-compatible output)
 - Agent CLI commands (build, list, show)
 - Integration adapter framework
 - MCP integration layer with server discovery

package/dist/index.d.ts

@@ -2322,7 +2322,7 @@ declare class StepTracker {
 /**
  * Skill category
  */
-type SkillCategory = 'core' | 'custom';
+type SkillCategory = 'core' | 'git' | 'jira' | 'docs' | 'custom';
 /**
  * Integration dependency for a skill
  */

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@deimoscloud/coreai",
-  "version": "0.1.19",
+  "version": "0.1.20",
   "description": "A configurable, team-ready AI agent orchestration platform",
   "type": "module",
   "main": "dist/index.js",

package/skills/core/sprint-status/SKILL.md

@@ -0,0 +1,65 @@
+---
+description: Review current sprint progress and status (EM only)
+argument-hint: [optional: sprint number or "current"]
+---
+
+# Sprint Status Review
+
+You are the Engineering Manager. Review the current sprint status and provide a summary.
+
+## Steps
+
+1. **Check Jira for current sprint:**
+   ```
+   Search: project = {{JIRA_PROJECT}} AND sprint in openSprints()
+   ```
+
+2. **Gather ticket status:**
+   - Count tickets by status (Backlog, In Progress, In Review, Done)
+   - Identify any blocked tickets
+   - Check for overdue items
+
+3. **Check agent inboxes for pending work:**
+   - `/KnowledgeLibrary/*/inbox/` - any unprocessed tasks?
+   - Are there completion summaries in your inbox you haven't processed?
+
+4. **Review open PRs:**
+   ```
+   gh pr list --state open
+   ```
+
+5. **Provide summary report:**
+
+```markdown
+## Sprint Status Report
+**Date:** [today]
+**Sprint:** [sprint name/number]
+
+### Ticket Summary
+| Status | Count | Tickets |
+|--------|-------|---------|
+| Done | X | {{JIRA_PROJECT}}-XX, {{JIRA_PROJECT}}-YY |
+| In Review | X | {{JIRA_PROJECT}}-XX |
+| In Progress | X | {{JIRA_PROJECT}}-XX |
+| Blocked | X | {{JIRA_PROJECT}}-XX (reason) |
+| Backlog | X | {{JIRA_PROJECT}}-XX |
+
+### Open PRs
+- PR #XX - [title] - [status: awaiting review/changes requested/approved]
+
+### Pending Agent Work
+- [agent]: [pending task or "inbox clear"]
+
+### Blockers & Risks
+- [Any blockers or risks to sprint completion]
+
+### Recommended Actions
+1. [Action item]
+2. [Action item]
+```
+
+## Additional Context
+
+$ARGUMENTS
+
+Now review the sprint status.

package/skills/docs/docs-update/SKILL.md

@@ -0,0 +1,201 @@
+---
+description: Use when updating documentation, adding to Confluence, documenting changes, updating local docs, or keeping documentation in sync. Handles both Confluence pages and local /docs files.
+argument-hint: <what to document> [where]
+requires: [documentation]
+---
+
+# Documentation Update
+
+You are documenting: **$ARGUMENTS**
+
+## Documentation Locations
+
+| Location | Content Type | When to Update |
+|----------|--------------|----------------|
+| **Confluence** | Architecture, PRD, standards, guides | Major decisions, feature specs, process changes |
+| **Local /docs** | Technical workflows, setup guides | Development process, CI/CD, environment setup |
+| **Code comments** | Implementation details | Complex logic, non-obvious code |
+| **README files** | Module overviews | New modules, significant changes |
+
+## Confluence Pages (Primary)
+
+### Update via Atlassian MCP
+
+#### Read Current Page
+
+```
+mcp__atlassian__getConfluencePage
+Page ID: <page-id>
+```
+
+#### Update Page
+
+```
+mcp__atlassian__updateConfluencePage
+Page ID: <page-id>
+Title: <page title>
+Content: <updated content in Confluence storage format>
+```
+
+#### Create New Page
+
+```
+mcp__atlassian__createConfluencePage
+Space Key: {{CONFLUENCE_SPACE}}
+Title: <page title>
+Content: <content in Confluence storage format>
+Parent Page ID: <parent-id> (optional)
+```
+
+### Confluence Content Format
+
+Confluence uses storage format (HTML-like). Common patterns:
+
+```html
+<h2>Section Title</h2>
+<p>Paragraph text.</p>
+
+<ul>
+  <li>Bullet point 1</li>
+  <li>Bullet point 2</li>
+</ul>
+
+<ac:structured-macro ac:name="code">
+  <ac:parameter ac:name="language">kotlin</ac:parameter>
+  <ac:plain-text-body><![CDATA[
+    // code here
+  ]]></ac:plain-text-body>
+</ac:structured-macro>
+
+<ac:structured-macro ac:name="info">
+  <ac:rich-text-body>
+    <p>Info box content</p>
+  </ac:rich-text-body>
+</ac:structured-macro>
+```
+
+## Local Documentation (/docs)
+
+### Update Local Docs
+
+Use the Edit tool to update existing documentation:
+
+```
+Edit file: {{PROJECT_ROOT}}/docs/[filename].md
+```
+
+### Documentation Template
+
+```markdown
+# [Title]
+
+**Last Updated:** [YYYY-MM-DD]
+**Author:** [agent-name]
+
+## Overview
+
+[Brief description of what this document covers]
+
+## [Main Section]
+
+[Content]
+
+### [Subsection]
+
+[Content]
+
+## Related Documentation
+
+- [Link to related doc 1]
+- [Link to related doc 2]
+
+---
+
+**Questions?** File a Jira ticket with component "Documentation".
+```
+
+## Keep Documentation in Sync
+
+When updating documentation, ensure both locations are updated:
+
+### Architecture Changes
+
+1. Update Confluence Architecture page (primary)
+2. Update `/KnowledgeLibrary/architecture.txt` (local cache)
+
+### Process Changes
+
+1. Update Confluence Development page (primary)
+2. Update `/docs/DEVELOPMENT_WORKFLOW.md` (local reference)
+
+### PRD Changes
+
+1. Update Confluence Product page (primary)
+2. Update `/KnowledgeLibrary/prd.txt` (local cache)
+
+## What to Document
+
+### Always Document
+
+- Architecture decisions and rationale
+- New modules or significant refactors
+- API changes or new endpoints
+- Configuration changes
+- Process changes
+- Breaking changes
+
+### Document When Significant
+
+- Bug fixes with non-obvious solutions
+- Performance optimizations
+- Security considerations
+- Integration details
+
+### Skip Documentation For
+
+- Minor bug fixes
+- Code formatting changes
+- Dependency version bumps (unless breaking)
+- Internal refactors with no API changes
+
+## Documentation Checklist
+
+Before completing documentation update:
+
+- [ ] Content is accurate and up-to-date
+- [ ] Examples are working and tested
+- [ ] Links are valid
+- [ ] Code snippets are correct
+- [ ] Related docs are cross-referenced
+- [ ] Both Confluence and local docs updated (if applicable)
+
+## Error Handling
+
+**If Confluence MCP is unavailable:**
+1. Update local documentation first
+2. Note in completion report: "Confluence update pending - please sync manually"
+3. Create a task to sync Confluence later
+
+**If page doesn't exist:**
+1. Create new page with `mcp__atlassian__createConfluencePage`
+2. Or create local doc and note Confluence page needs creation
+
+## Completion
+
+After documentation update, report:
+```
+Documentation Updated
+
+Updated:
+- [Confluence page name] (URL)
+- [Local file path]
+
+Changes:
+- [Summary of what was documented]
+
+Sync Status:
+- Confluence: [Updated/Pending]
+- Local docs: [Updated/N/A]
+```
+
+Now proceed with the documentation update.

package/skills/git/git-commit/SKILL.md

@@ -0,0 +1,145 @@
+---
+description: Use when committing code changes, pushing to remote, or saving work. This is a GATED commit that enforces quality checks - all configured checks must pass before the commit proceeds.
+argument-hint: [commit message or ticket reference]
+requires: [git]
+---
+
+# Gated Git Commit
+
+You are committing changes for: **$ARGUMENTS**
+
+## CRITICAL: Quality Gate
+
+**NO COMMIT will proceed until ALL checks pass.** This is non-negotiable.
+
+## Pre-Commit Checks (MANDATORY)
+
+Run these checks IN ORDER. If ANY fail, STOP and fix before proceeding.
+
+### 1. Code Formatting
+
+```bash
+{{LINT_CMD}}
+```
+
+**If fails:**
+- Run the format/fix command if available
+- Review changes
+- Re-run the check
+
+### 2. Static Analysis
+
+```bash
+{{STATIC_ANALYSIS_CMD}}
+```
+
+**If fails:**
+- Review the report
+- Fix code smells and issues
+- Re-run the check
+
+### 3. Unit Tests
+
+```bash
+{{TEST_CMD}}
+```
+
+**If fails:**
+- Review test failures
+- Fix failing tests or update tests for new behavior
+- Re-run tests
+
+### 4. Build Verification
+
+```bash
+{{BUILD_CMD}}
+```
+
+**If fails:**
+- Fix compilation errors
+- Re-run build
+
+## All Checks Passed?
+
+Only proceed to commit if you see **BUILD SUCCESSFUL** (or equivalent) for ALL checks above.
+
+## Commit Process
+
+### 1. Stage Changes
+
+```bash
+git add -A
+```
+
+Or stage specific files:
+```bash
+git add [specific files]
+```
+
+### 2. Commit with Proper Format
+
+**Commit message format:**
+```
+<type>(<scope>): <description> [{{JIRA_PROJECT}}-XX]
+
+[optional body with details]
+```
+
+**Types:** `feat`, `fix`, `refactor`, `test`, `docs`, `chore`, `style`, `perf`
+
+**Examples:**
+```bash
+git commit -m "feat(core): add new feature [{{JIRA_PROJECT}}-123]"
+git commit -m "fix(sync): resolve connection timeout [{{JIRA_PROJECT}}-456]"
+git commit -m "test(core): add unit tests for filtering [{{JIRA_PROJECT}}-789]"
+```
+
+### 3. Push to Remote
+
+```bash
+git push -u origin $(git branch --show-current)
+```
+
+## Commit Checklist
+
+Before committing, verify:
+- [ ] All quality checks passed
+- [ ] Commit message includes ticket number [{{JIRA_PROJECT}}-XX]
+- [ ] Commit message follows type(scope): description format
+- [ ] No debug code or commented-out blocks
+- [ ] No secrets or API keys in code
+- [ ] Changes are related to the ticket being worked on
+
+## Error Recovery
+
+**If checks fail repeatedly:**
+1. Report which specific check is failing
+2. Show the error output
+3. Ask for guidance if the fix is unclear
+
+**If push fails:**
+```bash
+# If remote has new commits, rebase first
+git fetch origin
+git rebase origin/main
+# Then push again
+git push -u origin $(git branch --show-current)
+```
+
+## Completion
+
+After successful commit and push, report:
+```
+Commit successful
+
+Branch: feature/{{JIRA_PROJECT}}-XX-description
+Commit: [short SHA] <commit message>
+Pushed to: origin/feature/{{JIRA_PROJECT}}-XX-description
+
+Quality checks passed:
+- [List of checks that passed]
+
+Ready for PR creation when implementation is complete.
+```
+
+Now proceed with the gated commit process.