alexbarakov

BI Strategy Self Consulting Agent — AI skill for Claude Code

AI community

Self-consulting agent for BI+AI strategy: maturity diagnosis, 7-block strategy wiki, 66-atom knowledge base, failure catalog, golden set.

How to install BI Strategy Self Consulting Agent

This entry records only its repository, not the path inside it, so there is no exact command to give. Open alexbarakov/bi-ai-strategy-builder and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What BI Strategy Self Consulting Agent does

Self-consulting agent for BI+AI strategy: maturity diagnosis, 7-block strategy wiki, 66-atom knowledge base, failure catalog, golden set. Pairs with DG-strategy-self-consulting-agent.

Alternatives in AI

RAG Skill — A Skill dedicated to local knowledge base retrieval 712 ★
Karpathy LLM Wiki — Agent Skills-compatible LLM wiki for Claude Code, Cursor, and Codex 618 ★
LLM Wiki Compiler — Claude Code plugin that compiles markdown knowledge files into a topic-based wiki 224 ★

README

BI Strategy — self-consulting agent

A ready-to-run skill that builds a **BI+AI strategy** for a specific company: interview → maturity diagnosis → a linked mini-wiki covering every block. Grounded on the canonical method (BI Project Health Check + D&A Strategy & Tactics Planner, data nature / Alex Barakov) plus the AI layer of the «BI+AI Strategy 2026» course.

Working principle: **AI drafts — humans validate.** The skill does not invent; it instantiates known frameworks against the participant's own data, and every block is validated by the user before the next one starts.

**Companion:** [DG Board KB](https://github.com/alexbarakov/DG-strategy-self-consulting-agent) — the governance half of the same method. The two repositories are designed as a pair and share the same invariants.

At a glance

What	Size
Knowledge base — one topic per file, `[[wiki-links]]`, machine-readable graph	66 atoms
Theme graph — nodes, clusters, curated relations, Mermaid views	66 nodes · 283 links · 10 diagrams
Participant question base with answers, provenance-tagged	101 Q&A
Golden set — retrieval / answer quality / honest refusal	143 positions
Failure catalog — named failure modes with symptom triage	75 in 7 families
Numbers registry, every figure tagged by reliability	93 rows
Health Check diagnostic factors, 0–4 scale	72 factors
Document structure	6 sections
Analysis frame behind it	7 blocks
Working templates (xlsx), ready to hand to the participant	2

Two ways to use this repository

**A. As a knowledge source (passive).** Add the repo or its URL to your agent's context and let it ground *other* tasks — answering BI/AI questions, reviewing a design, preparing a talk, sanity-checking a vendor claim. Nothing to install; the agent follows "How the agent uses it" below.

**B. As a strategy skill (active).** [`SKILL.md`](SKILL.md) runs the full procedure in three sce