boundary-bbcr-fallback
Execute automatic BBCR (Collapse-Rebirth Correction) when knowledge boundaries are exceeded or reasoning fails.
Systematic approach for building complex features incrementally, preventing premature completion and ensuring comprehensive functionality through structured feature tracking.
$ npx -y skills add qdhenry/Claude-Command-Suite --agent claude-codeHow it fires
How this command gets triggered: by you, by Claude, or both.
/incremental-feature-buildContext preview
What this command does when you run it.
Systematic approach for building complex features incrementally, preventing premature completion and ensuring comprehensive functionality through structured feature tracking.
Systematic approach for building complex features incrementally, preventing premature completion and ensuring comprehensive functionality through structured feature tracking.
Build a feature incrementally using the structured approach for: **$ARGUMENTS**
This command implements best practices for long-running agent tasks, preventing the tendency to one-shot applications or prematurely consider projects complete.
> **Note:** `$ARGUMENTS` is automatically replaced with the text following the command invocation. > Example: `/dev:incremental-feature-build user authentication system` sets `$ARGUMENTS` to "user authentication system"
---
# Create tracking directory with error handling
mkdir -p .feature-tracking || { echo "ERROR: Cannot create .feature-tracking directory. Check permissions."; exit 1; }If directory creation fails, verify:
Analyze the user's request and expand it into a comprehensive list of granular features. Each feature should be a discrete, testable unit of functionality.
**Create file: `.feature-tracking/features.json`**
{
"project": "$ARGUMENTS",
"created": "YYYY-MM-DD",
"version": "1.0.0",
"summary": {
"total": 0,
"passing": 0,
"failing": 0
},
"features": []
}Each feature MUST follow this exact JSON schema:
{
"id": "FEAT-001",
"category": "functional|ui|integration|performance|security|accessibility",
"priority": "critical|high|medium|low",
"description": "Clear, actionable description of the feature",
"steps": [
"Step 1: Specific action to verify",
"Step 2: Expected behavior check",
"Step 3: Edge case validation"
],
"dependencies": ["FEAT-000"],
"passes": false,
"implementedAt": null,
"commitHash": null
}> **Note:** This command uses a simple boolean `passes` field since features are implemented sequentially. > The parallel command (`/dev:parallel-feature-build`) uses a multi-state `status` field to track > concurrent work: `pending`, `in_progress`, `passed`, `blocked`.
Generate features across ALL relevant categories:
| Category | Description | Examples | |----------|-------------|----------| | `functional` | Core business logic | User can submit form, data saves correctly | | `ui` | User interface elements | Button displays, modal opens, responsive layout | | `integration` | System connections | API calls work, database syncs, auth flows | | `performance` | Speed and efficiency | Page loads under 3s, lazy loading works | | `security` | Protection measures | Input sanitized, auth required, CSRF protection | | `accessibility` | Inclusive design | Screen reader support, keyboard navigation |
When expanding the user's request:
1. **Use Hierarchical Organization**: Group features into epics/modules for manageability
2. **Be Specific**: Each feature should be independently verifiable 3. **Include Edge Cases**: Error states, empty states, boundary conditions 4. **Cover All Paths**: Happy path AND unhappy paths 5. **Think Like a User**: What would they expect at each step?
**Feature Grouping Example:**
{
"epics": [
{
"id": "EPIC-01",
"name": "User Authentication",
"features": ["FEAT-001", "FEAT-002", "FEAT-003"]
}
]
}---
**Create file: `.feature-tracking/PROGRESS.md`**
# Feature Implementation Progress ## Project: $ARGUMENTS ## Started: YYYY-MM-DD --- ## Current Status - **Features Total**: X - **Implemented**: 0 - **Remaining**: X --- ## Session Log ### Session 1 - YYYY-MM-DD **Focus**: Initial setup and first features **Completed**: - (none yet) **Notes**: - (session notes here) **Commit**: (hash when committed) --- ## Implementation Order Based on dependencies, implement in this order: 1. [ ] FEAT-001 - Description 2. [ ] FEAT-002 - Description ... --- ## Blocked/Issues (Track any blockers here)
Ensure git is initialized and create initial tracking commit:
git add .feature-tracking/ git commit -m "feat: initialize feature tracking for $ARGUMENTS - Created features.json with X features to implement - All features marked as failing (passes: false) - Created progress tracking file"
---
**You MUST follow these rules without exception:**
1. **ONE FEATURE AT A TIME**: Never work on multiple features simultaneously 2. **NO TEST REMOVAL**: It is **UNACCEPTABLE** to remove or edit features from features.json because this could lead to missing or buggy functionality 3. **ONLY CHANGE STATUS**: The only modification allowed to features.json is changing `passes` from `false` to `true` and updating `implementedAt` and `commitHash` 4. **COMMIT AFTER EACH**: Create a git commit after each feature is implemented 5. **CLEAN STATE**: Each commit must leave the codebase in a working state
For each feature, follow this exact workflow:
1. Read `.feature-tracking/features.json` 2. Find the first feature where `passes: false` 3. Check dependencies are satisfied (all dependent features pass) 4. Announce: "Now implementing FEAT-XXX: [description]"
1. Write the necessary code 2. Ensure existing functionality still works 3. Test the specific feature manuall
A comprehensive development toolkit designed following Anthropic's Claude Code Best Practices for AI-assisted software development.
Repo: qdhenry/Claude-Command-Suite
Execute automatic BBCR (Collapse-Rebirth Correction) when knowledge boundaries are exceeded or reasoning fails.
Analyze semantic position relative to knowledge boundaries to prevent hallucination and identify uncertainty zones.
Generate a visual heatmap of knowledge boundaries showing safe zones, risk areas, and semantic coverage.
Evaluate the current risk level and provide detailed analysis of potential hallucination or reasoning failure.
Find and construct semantic bridges to safely navigate from current position to target concept without crossing dangerous boundaries.
Takes an input prompt and returns ONLY a token-optimized version that preserves meaning while minimizing token count. Based on LLM tokenization principles:…