Run automated health checks for stale pages, orphan pages, and dead links.
npx skills add https://github.com/ltdrew/Agent-brain-2 --skill maintain-ltdrew
Please help me install this Agent Skill. Skill: maintain Source: https://github.com/ltdrew/Agent-brain-2/tree/main/skills/maintain Command: npx skills add https://github.com/ltdrew/Agent-brain-2 --skill maintain-ltdrew
Maintains brain health and data integrity by performing automated checks like backlinks, citations, stale content, and orphan pages, and by auditing and repairing the knowledge graph.
Run gbrain maintain to perform a full health check and fix identified issues.
High-intent search queries and answers about installing and using this skill.
To detect orphan pages and stale content in your knowledge base, run an automated health check that scans frontmatter metadata to flag isolated notes lacking inbound connections and outdated information. This process identifies stale pages and orphans during scheduled audits.
Backlink enforcement and citation audit for data integrity is the process of verifying that timeline references and provenance links exist across your graph. It ensures proper cross-references are maintained, preventing broken references and dead links within the knowledge base.
Populate and update a link graph and timeline automatically by running a structured maintenance workflow that scans cross-references and frontmatter triggers. This refreshes the brain's memory, populating missing links and updating the timeline graph across the knowledge base.
Yes, this knowledge graph maintenance approach works after large data imports by running a full maintenance cycle to refresh memory and fix inconsistencies. It repairs the link graph, enforces backlinks, and audits citations to restore data integrity.
Limitations of automated brain maintenance for stale pages include its reliance on structured frontmatter metadata, defined triggers, and tools to load context. Without proper metadata inputs, the automated checks for dead links and cross-references cannot execute effectively.
AradotsoDefine standard Agent Skills with SKILL.md frontmatter for cross-platform loading.
ngochuy13Audit and repair knowledge graph integrity across stale content, orphan pages, and dead links.
cncoderAudit Claude Code configuration files for consistency across CLAUDE.md, settings.json, agents, and rules.
NinatuziMaintains knowledge graphs by enforcing link integrity and validating data structures.
