Load methodology context for creating or fixing Claude Code instruction artifacts (discovery: implementation-clarity). Evaluate at implementation-clarity when artifact type is known (command, skill, hook, protocol, external tool config) and implementation approach needs methodology guidance.
/plugin marketplace add MariusWilsch/clarity-workflow-plugin/plugin install clarity-workflow@clarity-workflow-marketplaceThis skill inherits all available tools. When active, it can use any tool Claude has access to.
conversation_data/filtered_minimal_d13784a0-ec98-40c5-bd90-73e2fa1ca00b.jsonlreferences/command.mdreferences/external-tool-config.mdreferences/persuasion-principles.mdreferences/protocol.mdscripts/extract_conversation.pyscripts/monitor_session.shProvides methodology context for creating or fixing instruction artifacts.
After rubber-duck diagnoses the artifact type, load the appropriate methodology:
| Artifact Type | Reference Files | Fix Tool |
|---|---|---|
| Command/Prompt | command.md + persuasion-principles.md | /stage |
| Skill | skill.md + persuasion-principles.md | skill-creator |
| Hook | (not yet implemented) | settings.json |
| Protocol | protocol.md + persuasion-principles.md | CLAUDE.md edit |
| External Tool Config | external-tool-config.md | hippocampus |
Note: Command/Prompt covers ALL prompt types: slash commands, agent prompts, prompt templates. Any engineered prompt follows the same methodology.
Note: Hooks use settings.json (declarative config) - no persuasion principles needed. External tools use hippocampus for documentation - no persuasion principles needed. Commands, skills, and protocols require persuasion techniques to ensure AI compliance.
After implementing the fix, do NOT commit immediately. Follow this flow:
Announce: "Fix implemented. Verify in a new project session before committing."
After user provides Session C conversation path, START MONITORING IMMEDIATELY.
User providing a path = start monitoring loop. No confirmation needed.
When user provides a Session C path, YOU MUST start the monitoring loop:
sleep 30 && ~/.claude/skills/manage-artifact/scripts/monitor_session.sh "<path>" 5
Repeat this cycle until you see verification results (success or failure). Do NOT do a single extraction and stop - keep watching Session C progress in real-time.
Script arguments:
path: Conversation .jsonl file path (required)num_messages: Messages to display (default: 5)Based on monitoring results:
Command, Skill, Protocol, and External Tool Config are supported. Hook support will be added when needed.
For commands/prompts (slash commands, agent prompts, templates):
For skills:
For protocols:
For external tool config:
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.