One-click install
npx skills add https://github.com/Lambenthan/empiricalwiki --skill reset-lambenthan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: reset
Source: https://github.com/Lambenthan/empiricalwiki/tree/main/.claude/skills/reset
Command: npx skills add https://github.com/Lambenthan/empiricalwiki --skill reset-lambenthan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the need to recover from broken or experimental wiki states by resetting EmpiricalWiki back to a clean scaffold without having to manually delete many files.

Core Features & Use Cases

  • Scoped destructive reset: Resets only selected areas (wiki, raw, log, checkpoints, or all) to minimize unintended loss.
  • Dry-run planning: Generates a JSON deletion/reset plan first, so you can verify what will change.
  • Guardrails for irreversibility: Requires explicit confirmation and warns that raw data deletion is irreversible.
  • Use case: You run an ingestion experiment and the resulting wiki graph/log/checkpoints become inconsistent; use the appropriate scope to cleanly restart ingestion and research.

Quick Start

Run /reset with the specific scope you want to rebuild, for example: /reset --scope wiki.

Frequently Asked Questions about reset

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I rebuild an EmpiricalWiki workspace after a failed ingestion experiment?

To rebuild an EmpiricalWiki workspace, you run a scoped destructive reset that deletes or re-initializes wiki, raw, logs, and checkpoints based on your selected scope, restoring a clean scaffold for restarting ingestion and research.

Can I preview what files will be deleted before resetting my wiki workspace?

Yes, you can preview file deletions by using the dry-run planning feature, which generates a JSON deletion and reset plan first so you can verify exactly what will change before executing destructive file operations.

What is the best way to clean up inconsistent wiki graphs and checkpoints without losing all workspace data?

The best way to clean up inconsistent wiki graphs is using a scoped reset, targeting only specific areas like wiki or checkpoints rather than all, which minimizes unintended data loss while restoring consistency.

How do I safely perform a destructive workspace recovery on EmpiricalWiki?

Safe destructive workspace recovery requires a mandatory scope parameter and explicit confirmation, warning that raw data deletion is irreversible while preserving required placeholders like .gitkeep and wiki/CLAUDE.md during re-initialization.

Does the reset process preserve essential placeholder files like .gitkeep when re-initializing the workspace?

Yes, the reset process preserves required placeholder files like .gitkeep and wiki/CLAUDE.md during re-initialization, ensuring the workspace scaffold remains structurally intact after destructive cleanup operations complete.

When should I not use a full workspace reset for my knowledge ingestion workflow?

You should avoid a full workspace reset when only specific areas need correction, as raw data deletion is irreversible; instead, use targeted scopes like wiki or checkpoints to minimize unintended loss of unrelated data.