Bloomin Openclaw

作成者 @sidd791

Bloomin OpenClaw agent configuration — SOUL, skills, memory, and routing rules for the BloomBrain AI agent powered by OpenClaw.

ClawLodge CLI でインストールclawlodge-cli が必要です
npm install -g clawlodge-cli
clawlodge install sidd791-bloomin-openclaw
向いている用途
内容
18 ファイル数v0.1.0

Bloomin OpenClaw

Agent configuration for BloomBrain — Bloomin's internal AI assistant powered by OpenClaw.

Summary

Configuration workspace defining BloomBrain's identity, knowledge, skills, and tool integrations. BloomBrain operates in Slack and has access to brand intelligence, competitor research, Reddit ICP data, and marketing automation tools.

Included Assets

Core Configuration

  • SOUL.md — Identity, brand knowledge, core mechanism (UMS), market priorities
  • AGENTS.md — Session startup workflow and memory management rules
  • MEMORY.md — Long-term curated memory (main session only)
  • TOOLS.md — MCP tool references, Slack channels, repo paths

Daily Memory

  • memory/YYYY-MM-DD.md — Session logs and operational notes
  • memory/bloomin-UMS.md — Unique Mechanism Statement and brand strategy
  • memory/bloomin-brand.md — Product details, ingredients, positioning
  • memory/bloomin-gummies.md — Gummy product strategy and hypotheses
  • memory/bloomin-personas-*.md — Customer personas (P4–P11 documented)
  • memory/competitor*.md — Competitor intelligence and Trustpilot review mining

Competitor Research

  • memory/competitors/INDEX.md — Active competitor tracking index
  • memory/competitors/*.md — Individual competitor profiles (HerSolution, Vigorelle, Unfabled, BrainMD Happy Saffron, TikTok gummy brand)

Reddit Intelligence System

  • reddit-intelligence/ — Saad's Reddit scraping and ICP analysis pipeline
  • reddit-intelligence/system/ — API credentials, architecture docs, Supabase schemas
  • reddit-intelligence/intelligence/ — Live ICP profiles and copy vault
  • reddit-intelligence/skills/ — Planned automation skills for Reddit intel

How to Use

This workspace is loaded by BloomBrain on session startup. Follow the startup sequence in AGENTS.md:

  1. Read SOUL.md for identity and brand context
  2. Read USER.md for the current user's context
  3. Load today's and yesterday's daily notes from memory/
  4. In main sessions: also load MEMORY.md

BloomBrain responds to Slack mentions (@BloomBrain) in channels:

  • the-bloominbrain (C0ALRNYBJJG) — interactive queries
  • the-bloominbrain-reports (C0AMVG202KC) — automated reports

Notes

  • Markets: USA primary; UK, Australia, Canada secondary. EU data appears in research for pattern recognition only.
  • Primary language: English. Dutch supported for NL market.
  • Security: MEMORY.md loads only in direct/main sessions — never in shared contexts.
  • UMS: All brand decisions trace back to the "Survival Mode" mechanism — read memory/bloomin-UMS.md for copy and strategy foundations.
  • Data freshness: Reddit intelligence updated regularly via Saad's pipeline. Competitor reviews mined via Trustpilot scraper.

Workspace

Updated 2026-04-03 18:06:31Published via clawlodge-cli/0.1.8
AGENTS.md
text · 7.9 KB

Loading preview...

No comments yet.

関連Lobster