wiki-aggregate

Aggregate 3–30 research artifacts into a structured pack with provenance.

231|24|Updated Apr 11, 2026

Search Tags:#research #provenance #evidence #synthesis #aggregate #cross-source

Ready to run

npx skills add https://github.com/LichAmnesia/lich-skills --skill wiki-aggregate

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: wiki-aggregate
Source: https://github.com/LichAmnesia/lich-skills/tree/main/skills/wiki-aggregate
Command: npx skills add https://github.com/LichAmnesia/lich-skills --skill wiki-aggregate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Aggregates N raw research artifacts into a single structured pack with cross-source provenance so you don't lose intermediate evidence in summaries.

Core Features & Use Cases

Build a provenance-rich pack by navigating sources with lightweight tools (inspect / search / synthesize) instead of concatenating content.
Output a complete pack: brief.md, findings.md with claims and evidence, sources.tsv, and an aggregation log for auditability.
Supports 3–30 sources, surfaces cross-source contradictions, and tracks subtopics until coverage criteria are met.
Maintains a reproducible audit trail for later re-analysis.

Quick Start

Aggregate your 3–30 raw notes into a single structured pack by using the lite-aggregator loop and review the generated brief.md, findings.md, sources.tsv, and _aggregation_log.md.

Frequently Asked Questions about wiki-aggregate

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

FAQPage Schema

What is cross-source provenance tracking for research aggregation?▼

Cross-source provenance tracking records the exact origin of evidence when aggregating multiple research artifacts. It uses path:line references in a notes scratchpad so you never lose intermediate evidence during synthesis.

How do I aggregate multiple research notes into a structured pack?▼

To aggregate research notes, use a lite agent workflow with inspect, search, and synthesize tools to extract evidence from 3–30 sources. It generates a structured pack containing brief.md, findings.md, sources.tsv, and an aggregation log.

Can I use this aggregation method for transcripts and raw analyses?▼

Yes, this aggregation method supports 3–30 raw notes, transcripts, or analyses. It applies cross-source consistency checks and surfaces contradictions across all supported source types until coverage criteria are met.

What's the best way to maintain an audit trail during research synthesis?▼

The best way to maintain an audit trail is to generate an _aggregation_log.md alongside your findings. This tracks subtopics and provenance entries, providing a reproducible record for later re-analysis of your evidence pack.

Why does my research synthesis lose intermediate evidence when summarizing?▼

Summarizing loses intermediate evidence because it concatenates content instead of tracking origins. This aggregation method avoids that by using lightweight tools to extract evidence and build a notes scratchpad with path:line provenance.

Does this cross-source aggregation work without external dependencies?▼

Yes, cross-source aggregation requires no external dependencies. It operates as a lite agent workflow using internal inspect, search, and synthesize tools to process your artifacts and build the structured output pack.