Skip to content
Conduit

skill

brand-soul

Brand Soul — an open-source framework created by Cacao Web Studio for persistent, evidence-aware AI brand identity.

CacaoWebStudio/brand-soulcommunitylistedunreviewed

Set up

Skills are installed by placing their folder where your agent loads skills.

The download comes from github.com/CacaoWebStudio/brand-soul itself, not from our server, so what you get is exactly what its author published.

Install instructions come from the README

Skills do not use an MCP config. Follow the steps in the README below.

README

synced Oct 23, 2026

Apache 2.0 Agent Skill Codex Claude Gemini Grok

Give every AI agent the same durable understanding of your brand.


Brand Soul — an open-source framework created by Cacao Web Studio

An open-source framework and Codex Skill for creating and maintaining a brand’s persistent AI identity.

Created and maintained by Cacao Web Studio.

Brand Soul discovers or helps define what makes a brand non-interchangeable, records the evidence and human decisions behind it, and turns that knowledge into a portable Source of Truth. Future AI agents can use the same identity without reinterpreting the brand from scratch.

Brand Soul is not a tone-of-voice generator or an AI mythology machine. It prefers evidence over marketing, decisions over adjectives, tensions over fake consistency, and questions over invention.

Why Brand Soul?

Most AI workflows begin by reconstructing a brand from scattered websites, old decks, prompts, and assumptions. The result drifts from channel to channel—and unsupported stories quietly become “facts.”

Brand Soul creates a governed layer between the real brand and every AI system acting on its behalf:

flowchart LR
    E["Evidence"] --> BS["Brand Soul"]
    H["Founder decisions"] --> BS
    BS --> T["Truth"]
    BS --> I["Identity"]
    BS --> S["Strategy"]
    BS --> X["Expression"]
    T & I & S & X --> C["Brand Context Contract"]
    C --> A["Ads agent"]
    C --> W["Web agent"]
    C --> M["Marketing agent"]
    C --> U["Support agent"]

Main capabilities

CapabilityWhat it does
DiscoverInspects evidence before interviewing or writing brand copy
ExcavateSeparates enduring identity from old positioning and historical noise
CreateHelps founders make explicit identity choices for new brands
StructureProduces a compact human- and machine-readable Brand Source of Truth
ProtectGoverns facts, claims, contradictions, boundaries, and founder approvals
EvolveUpdates strategy without silently rewriting truth or protected identity
AuditChecks factual, story, identity, voice, cultural, positioning, and decision integrity
Stay distinctiveRejects identity language that a competitor could use unchanged

One framework, many AI agents

The technical Skill name is brand-soul. The repository follows the portable Agent Skills structure and works natively with current Agent Skills implementations in Codex, Claude Code, Gemini CLI, and Grok. Any document-capable AI can also use it by reading SKILL.md and the routed references.

AgentSupportTypical installation
OpenAI CodexNative Skill~/.codex/skills/brand-soul
Claude CodeNative Agent Skill~/.claude/skills/brand-soul
Gemini CLINative Agent Skillgemini skills install https://github.com/CacaoWebStudio/brand-soul
GrokNative Skill~/.grok/skills/brand-soul
Other AI agentsPortable instructionsExpose the repository and direct the agent to AGENTS.md or SKILL.md

See AI agent compatibility for invocation paths and portability boundaries.

Install

Codex

git clone https://github.com/CacaoWebStudio/brand-soul.git ~/.codex/skills/brand-soul

Then ask: Use $brand-soul to build a Brand Source of Truth for my company.

Claude Code

git clone https://github.com/CacaoWebStudio/brand-soul.git ~/.claude/skills/brand-soul

Then ask naturally or invoke /brand-soul.

Gemini CLI

gemini skills install https://github.com/CacaoWebStudio/brand-soul

Approve activation when Gemini asks, then use /skills list to confirm installation.

Grok

git clone https://github.com/CacaoWebStudio/brand-soul.git ~/.grok/skills/brand-soul

Then ask naturally or invoke /brand-soul.

Review any third-party Skill before installation. Brand Soul uses network access only for an optional GitHub Release check and a user-authorized update.

Updates

Brand Soul checks GitHub Releases at most once every 24 hours when an agent session can run Python with network access. The check is read-only, cached, non-blocking, and silent unless a newer stable release exists.

Check manually:

python3 scripts/check_for_updates.py

Install the latest stable release interactively:

python3 scripts/update_skill.py

Opt into a non-interactive update for a trusted scheduler or startup task:

python3 scripts/update_skill.py --auto

The updater requires a clean Git clone of the official repository, targets the tag attached to the latest published stable release, and uses a fast-forward merge. It refuses dirty installations, unofficial remotes, divergent history, and automatic major-version upgrades. Start a new agent session after updating so the new instructions are loaded.

For reproducible projects, keep a version pinned and update deliberately. To receive GitHub notifications, select Watch → Custom → Releases on this repository.

Maintainers publish updates in this order:

  1. Merge tested changes to main.
  2. Update both VERSION and the metadata.version field in SKILL.md.
  3. Tag the exact release commit as vX.Y.Z.
  4. Publish a GitHub Release from that tag with meaningful release notes.

The update checker intentionally follows the latest stable GitHub Release, not untagged commits on main.

How it works

Brand Soul operates in four modes:

  1. Build — discover an existing brand or facilitate deliberate choices for a new one.
  2. Update — classify new information, preserve history, and route protected changes through approval.
  3. Audit — test brand-facing work against ten integrity dimensions.
  4. Consume — let another agent load only the approved context it needs.

The generated repository stays intentionally compact:

brand-soul-<brand>/
├── brand-context.yaml       # machine entrypoint and consumer contract
├── truth.yaml               # atomic, evidence-linked facts
├── identity.md              # protected story, principles, tensions, boundaries
├── voice.md                 # observable language behavior
├── strategy.md              # current, challengeable strategic choices
└── governance/
    ├── evidence.yaml        # source register
    ├── claims.yaml          # repeatability permissions
    ├── issues.yaml          # proposals, gaps, contradictions
    └── decisions/           # rationale for material changes

Quick start

Ask your installed agent to use Brand Soul, or initialize a blank repository directly:

python3 scripts/initialize_brand_repository.py ./workspace \
  --brand-name "Example Brand" \
  --founder "Founder Name"

Validate it at any stage:

python3 scripts/validate_brand_repository.py ./workspace/brand-soul-example-brand

A draft with visible gaps can still be structurally valid. Approval requires an explicit founder review and a SHA-256 binding to the exact protected identity file.

Integrity model

Brand Soul audits ten dimensions:

  • Fact integrity
  • Identity integrity
  • Story integrity
  • Claim integrity
  • Boundary integrity
  • Cultural integrity
  • Voice integrity
  • Positioning integrity
  • Decision integrity
  • Distinctiveness

The deterministic validator checks structure, paths, states, evidence references, and approval hashes. Semantic checks remain reasoning-based because a schema cannot decide whether a story is true or a brand is distinctive.

Repository map

SKILL.md                     Canonical agent instructions
references/                  Methodology, contract, governance, evaluation
assets/                      Generated Brand Soul repository template
scripts/                     Dependency-free initialization, validation, and safe update utilities
evals/                       Behavioral fixtures and structural tests
agents/openai.yaml           Optional Codex UI metadata
AGENTS.md                    Portable fallback entrypoint

Development

Brand Soul uses Python’s standard library and has no required package installation.

python3 evals/test_structural.py
python3 evals/test_updates.py
python3 scripts/validate_brand_repository.py assets/brand-repository-template

See CONTRIBUTING.md before proposing changes. Contributions should improve evidence integrity, portability, governance, or measurable behavior—not add generic brand advice or channel-specific execution.

Project stewardship

Brand Soul is a public open-source project developed, branded, and maintained by Cacao Web Studio.

License

Licensed under the Apache License 2.0. Copyright 2026 Cacao Web Studio.

Apache-2.0 permits commercial and private use, modification, and distribution under its terms. Cacao Web Studio’s names and marks are not granted for unrestricted trademark use by the license.

↑↓ navigate openesc close