Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions .beads/issues.jsonl

Large diffs are not rendered by default.

470 changes: 459 additions & 11 deletions README.md

Large diffs are not rendered by default.

6 changes: 3 additions & 3 deletions docs/ai-development-best-practices.md
Original file line number Diff line number Diff line change
Expand Up @@ -56,7 +56,7 @@ A canonical, enumerated list of best practices derived from comprehensive analys

22. **Context Optimization**: Match AI tool selection to task complexity using [Context Optimization](../experiments/README.md#context-optimization) and defined criteria: simple models for queries <500 tokens, complex models for architecture decisions >2000 tokens.

23. **Image Spec**: Use diagrams, mockups, and flows as primary specifications for complex needs using [Image Spec](../experiments/README.md#image-spec), supplemented with structured text for precise details.
23. **Image Spec**: Use diagrams, mockups, and flows as primary specifications for complex needs using [Image Spec](../README.md#image-spec), supplemented with structured text for precise details.

24. **Version Knowledge Assets**: Treat AI development knowledge as code with semantic versioning, peer review processes, and automated quality checks.

Expand All @@ -66,9 +66,9 @@ A canonical, enumerated list of best practices derived from comprehensive analys

26. **Smart Defaults with Extensibility**: Design AI development tools that work perfectly with no configuration while supporting advanced customization options.

27. **Self-Documenting Commands**: Include comprehensive built-in help with usage patterns, examples, and parameter explanations in every AI development tool using [Custom Commands](../experiments/README.md#custom-commands).
27. **Self-Documenting Commands**: Include comprehensive built-in help with usage patterns, examples, and parameter explanations in every AI development tool using [Custom Commands](../README.md#custom-commands).

28. **Hooks-Based Governance**: Implement flexible hook systems for automated monitoring, logging, and policy enforcement at critical development workflow points using [Event Automation](../experiments/README.md#event-automation).
28. **Hooks-Based Governance**: Implement flexible hook systems for automated monitoring, logging, and policy enforcement at critical development workflow points using [Event Automation](../README.md#event-automation).

29. **Modular Command Architecture**: Organize AI tools into clear categories with well-defined scopes (project-specific vs. machine-wide) for flexible workflows.

Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -79,5 +79,5 @@ EOF

## Related Patterns

- [Spec-Driven Development](../../../README.md#spec-driven-development)
- [Automated Traceability](../../../README.md#automated-traceability)
- [Spec-Driven Development](../../README.md#spec-driven-development)
- [Automated Traceability](../../README.md#automated-traceability)
Original file line number Diff line number Diff line change
Expand Up @@ -92,5 +92,5 @@ Automatic approval if ALL conditions met:

## Related Patterns

- [Guided Refactoring](../../../README.md#guided-refactoring)
- [Testing Orchestration](../../README.md#testing-orchestration)
- [Guided Refactoring](../../README.md#guided-refactoring)
- [Testing Orchestration](../../experiments/README.md#testing-orchestration)
Original file line number Diff line number Diff line change
Expand Up @@ -95,5 +95,5 @@ if (!API_KEY) throw new Error("STRIPE_API_KEY required")

## Related Patterns

- [Security Orchestration](../../../README.md#security-orchestration)
- [Security Sandbox](../../../README.md#security-sandbox)
- [Security Orchestration](../../README.md#security-orchestration)
- [Security Sandbox](../../README.md#security-sandbox)
Original file line number Diff line number Diff line change
Expand Up @@ -104,5 +104,5 @@ git diff --name-only HEAD~1 HEAD

## Related Patterns

- [Testing Orchestration](../../README.md#testing-orchestration)
- [Suite Health](../../README.md#suite-health)
- [Testing Orchestration](../../experiments/README.md#testing-orchestration)
- [Suite Health](../../experiments/README.md#suite-health)
28 changes: 28 additions & 0 deletions examples/image-spec/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,28 @@
# Image Spec Example

This example provides a repeatable, image-first workflow for the [Image Spec](../../README.md#image-spec) pattern: use diagrams and mockups as primary specifications, then iterate with annotated visuals.

## Workflow

1. Create a **single** high-level diagram (architecture, UI mock, or flow) with labels and constraints (ports, component names, state transitions).
2. Attach the image and provide a short prompt that clarifies only what the image cannot (tech stack, scope, constraints).
3. Run the generated result, capture a screenshot, and annotate what’s missing or incorrect.
4. Re-attach the annotated image and iterate on one slice at a time.

## Recommended Image Set

- `architecture.png`: components + boundaries + ports
- `data-model.png`: fields + relationships + example payloads
- `ui-mock.png`: layout + key interactions
- `flow.png`: sequence of steps + decision points

## Prompt Template

```text
Build the system from the attached images.
Tech stack: <stack>
Scope: <what to implement first>
Constraints: <performance/security/testing constraints>
Output: code + tests + a short runbook
```

27 changes: 27 additions & 0 deletions examples/progressive-disclosure/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,27 @@
# Progressive Disclosure Example

This example provides a tool-agnostic starter layout for implementing the [Progressive Disclosure](../../README.md#progressive-disclosure) pattern in a real repository.

## Goal

Keep a small, universal rules file that applies to every task, and load specialized rules only when the work context requires them (security, API design, testing, CI/CD, etc.).

## Starter Layout

```text
.ai/
├── CLAUDE.md # Universal rules + routing hints (<60 lines)
├── rules/
│ ├── security/ # secrets, auth, dependencies
│ ├── development/ # api-design, database, testing
│ ├── operations/ # deployment, monitoring, cicd
│ └── architecture/ # patterns, performance
└── prompts/ # reusable task templates
```

## Suggested Workflow

1. Keep `.ai/CLAUDE.md` minimal and explicit about **when** to load specialized rules.
2. Add small, focused rule files under `.ai/rules/` (aim for <100 lines each).
3. Optionally combine with [Event Automation](../../README.md#event-automation) to auto-load the relevant rules before edits, writes, or test runs.

Loading
Loading