Lore
WarnAudited by ClawScan on May 10, 2026.
Overview
Lore is a coherent knowledge-base skill, but it tells the agent to persist conversation and external workspace content without clear user approval, scope, retention, or privacy limits.
Before installing, confirm where Lore data is stored, who can access it, how to delete it, and require the agent to ask before ingesting meeting notes, email, Slack/Notion/GitHub content, or other sensitive material.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent could add sensitive or unwanted content to Lore without the user realizing it first.
This directs autonomous use of a mutating ingestion tool whenever the agent judges content worth saving, without an explicit user approval step or clear limits on what may be stored.
Push content into Lore using the `ingest` tool whenever you encounter information worth preserving
Require explicit user confirmation before each ingestion, define allowed projects and sources, and provide clear skip/redaction rules for sensitive content.
Private conversations, emails, documents, or workspace records could become searchable and reusable in future tasks beyond the user's immediate request.
The skill instructs the agent to persist content from potentially private external systems into a reusable knowledge repository, but does not specify consent, retention, access boundaries, redaction, or reuse limits.
When you fetch content from Slack, Notion, GitHub, email, or other systems, ingest the relevant parts into Lore.
Only ingest user-approved content, restrict ingestion to clearly named sources and projects, redact secrets and personal data, and document retention and deletion controls.