Skip to content

QUICKSTART.md

Latest commit

 

History

History
104 lines (78 loc) · 3.34 KB

QUICKSTART.md

File metadata and controls

104 lines (78 loc) · 3.34 KB

Quickstart

This is the shortest path to test the approach on your own product.

1. Install the skills into Codex

cd /absolute/path/to/codex-second-brain-starter-kit
./scripts/install_codex_starter.sh

This copies the bundled generic skills into ${CODEX_HOME:-$HOME/.codex}/skills.

2. Initialize a multi-brain portfolio

Replace the placeholders below with your own values:

./scripts/second_brain_wizard.py init-portfolio \
  --portfolio-root "/absolute/path/to/second-brain-portfolio" \
  --name "Acme Product Portfolio" \
  --obsidian-root "/absolute/path/to/Obsidian/Product Brains"

What this creates:

  • a portfolio registry
  • shared roots for workspaces, manifests, and default vault locations
  • a repeatable starting point for multiple second brains

3. Add your first second brain

./scripts/second_brain_wizard.py add-brain \
  --portfolio-root "/absolute/path/to/second-brain-portfolio" \
  --name "Acme Platform" \
  --slug "acme-platform" \
  --mode hybrid

What this creates:

  • a starter manifest
  • a starter intelligence profile
  • a scaffolded Obsidian vault
  • a first vault audit
  • a first engineering-readiness report

4. Put real material into the workspace

Open the generated manifest and place your real source material in the paths it describes:

  • sources.corpus_path
  • repositories.local_clone_root

Then:

  • replace the sample repository entries with your actual repositories
  • update profile.intelligence_path if you want the profile elsewhere
  • edit the generated config/intelligence-profile.yaml to reflect your real capabilities, keywords, and repo mappings

5. Validate the manifest

python3 ./skills/product-intelligence-factory/scripts/validate_product_manifest.py \
  --manifest "/absolute/path/to/second-brain-portfolio/manifests/acme-platform.yaml" \
  --check-paths

6. Open Codex and use the starter prompts

Recommended order:

  1. prompts/01_bootstrap_product.md
  2. prompts/02_ingest_and_build_second_brain.md
  3. prompts/03_build_engineering_layer.md
  4. prompts/04_bug_workflow.md
  5. prompts/05_create_automations.md
  6. prompts/06_review_readiness.md

If you want the packaged helpers to do the heavy lifting before the prompt pass, run:

python3 ./skills/product-intelligence-factory/scripts/build_source_indices.py \
  --manifest "/absolute/path/to/second-brain-portfolio/manifests/acme-platform.yaml"

OPENAI_API_KEY="..." \
python3 ./skills/product-intelligence-factory/scripts/rebuild_product_brain.py \
  --manifest "/absolute/path/to/second-brain-portfolio/manifests/acme-platform.yaml"

7. Prove the workflow before scaling it

A good first test should end with:

  • a usable Obsidian home note
  • linked research and engineering notes
  • at least one code-intelligence path from docs to code
  • a real readiness report
  • a clear list of what is still manual

Helpful wizard commands

./scripts/second_brain_wizard.py list-brains --portfolio-root "/absolute/path/to/second-brain-portfolio"
./scripts/second_brain_wizard.py doctor --portfolio-root "/absolute/path/to/second-brain-portfolio"
./scripts/second_brain_wizard.py refresh --portfolio-root "/absolute/path/to/second-brain-portfolio"

refresh rebuilds source indices and vault notes before regenerating audit and readiness outputs. Use --metadata-only when you only want the audit/readiness pass.