Extract shared knowledge from 5 command files into 6 reusable skills: - plugin-discovery.md: Plugin scanning and discovery - interface-parsing.md: README.md and CLAUDE.md parsing - dependency-analysis.md: MCP server and data flow analysis - validation-rules.md: Compatibility and agent validation - mcp-tools-reference.md: Available MCP tools - visual-output.md: Standard formatting and headers Slim commands from 263-164 lines down to 44-55 lines each. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
1.9 KiB
1.9 KiB
Neurodivergent-Friendly Accommodations
Guidelines for making clarification interactions accessible and comfortable for neurodivergent users.
Core Principles
Reduce Cognitive Load
- Maximum 4 options per question
- Always include "Other" escape hatch
- Provide examples, not just descriptions
- Use numbered lists for easy reference
Support Working Memory
- Summarize frequently
- Reference earlier decisions explicitly
- Do not assume user remembers context from many turns ago
- Echo back understanding before proceeding
Allow Processing Time
- Do not rapid-fire questions
- Validate answers before moving on
- Offer to revisit or change earlier answers
- One question block at a time
Manage Overwhelm
- Offer to break into smaller sessions
- Prioritize must-haves vs nice-to-haves
- Provide "good enough for now" options
- Acknowledge complexity openly
Question Formatting Rules
Always do:
**How should errors be handled?**
1. Silent logging (user sees nothing)
2. Toast notifications (brief, dismissible)
3. Modal dialogs (requires user action)
4. Other
[Context: This affects both UX and error-handling complexity]
Never do:
How do you want to handle errors? There are many approaches...
Conflict Acknowledgment
Before asking about something that might conflict with a previous answer:
[Internal check]
Previous: User said "keep it simple"
Current question about: Adding configuration options
Potential conflict: More options = more complexity
Then acknowledge: "Earlier you mentioned keeping it simple. With that in mind..."
Escalation for Overwhelm
If the request is particularly complex or user seems overwhelmed:
- Acknowledge the complexity openly
- Offer to start with just ONE aspect
- Build incrementally
"This touches a lot of areas. Rather than tackle everything at once,
let's start with [most critical piece]. Once that's clear, we can
add the other parts. Sound good?"