brain-keeper

Ingest sources, update brain pages, and log provenance decisions.

49|7|Updated May 4, 2026
One-click install
npx skills add https://github.com/agencia-conversion/agentic-seo-skills --skill brain-keeper
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: brain-keeper
Source: https://github.com/agencia-conversion/agentic-seo-skills/tree/main/skills/brain-keeper
Command: npx skills add https://github.com/agencia-conversion/agentic-seo-skills --skill brain-keeper

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Brain Keeper helps you ingest sources and keep the project’s authorial “brain” consistent, decision-aware, and provenance-safe so every downstream agent can trust what’s recorded.

Core Features & Use Cases

  • Source ingestion and provenance logging: captures newly provided raw sources untouched and records the ingestion and its evidence in the log.
  • Brain page updates with guardrails: proposes and applies updates to canonical brain pages (e.g., identity, voice, technology, topic-clusters, review) only when backed by evidence and recorded as type: decision.
  • Linting for link integrity and editorial quality: runs editorial and structural checks (lexicon-driven) and blocks publishing-grade changes when required rules fail.
  • Content publication registration: registers newly published content into the canonical project/contents/ metadata and ties it to brain structures via clusters and evidence.

Quick Start

Use brain-keeper when you receive a new source or need to update a brain page so the change is logged with evidence and the remaining constraints are enforced.

Frequently Asked Questions about brain-keeper

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

FAQPage Schema
How do I log provenance for new sources ingested into a knowledge base?

To log provenance for ingested sources, you capture raw sources untouched and record the ingestion event along with its evidence in the decision log. This ensures every brain page update remains evidence-backed and decision-aware.

What is editorial linting for wikilinks and content registration?

Editorial linting for wikilinks is an automated check that enforces link resolution rules and structural constraints. It blocks publishing-grade changes when required references fail, preventing unsupported claims and broken links in the content brain.

How do I update a canonical brain page with evidence-backed decisions?

You update a canonical brain page by proposing changes only when supported by valid evidence. The update is then recorded as a decision type entry in the log, ensuring strict write boundaries and provenance safety are maintained.

Can I register newly published content and tie it to topic clusters?

Yes, you can register newly published content into canonical project metadata directories. This registration process ties the content to existing brain structures by mapping it to topic clusters and linking associated evidence.

Why does my brain page update get blocked by lint checks?

Your brain page update gets blocked by lint checks because editorial rules enforce strict write boundaries and wikilink resolution. It prevents unsupported claims or broken references from entering the canonical project directories.

How do I handle contradictions found during source ingestion?

To handle contradictions found during source ingestion, the system applies evidence-backed authorial updates and records the conflict resolution as a decision. This maintains consistency and provenance safety across the project brain.