base120

Consuming Base120 v1.0.0

Status: Canonical, Frozen, Audit-Clean
Trust Anchor: Sealed seed with SHA-256 + MRCC binding

Overview

Base120 v1.0.0 is a deterministic governance substrate containing 120 canonical mental models across 6 domains. This document describes how downstream systems should consume Base120 as immutable infrastructure.

Trust Surface

The following artifacts constitute the complete trust surface:

Artifact Path Purpose
Seed JSON artifacts/base120.v1.0.0.seed.json Canonical 120-model corpus
SHA-256 Hash artifacts/base120.v1.0.0.seed.sha256 Integrity verification
MRCC Binding compliance/base120.v1.0.0.seed.mrcc.json Governance claims and provenance

Verification (Required)

Before consuming Base120, downstream systems MUST verify integrity:

Python

import json
import hashlib
from pathlib import Path

def verify_base120_seed(base120_path: Path) -> bool:
    """Verify Base120 seed integrity. Returns True if valid."""
    seed_path = base120_path / "artifacts" / "base120.v1.0.0.seed.json"
    hash_path = base120_path / "artifacts" / "base120.v1.0.0.seed.sha256"
    mrcc_path = base120_path / "compliance" / "base120.v1.0.0.seed.mrcc.json"
    
    # Compute hash
    with open(seed_path, "rb") as f:
        computed = hashlib.sha256(f.read()).hexdigest()
    
    # Read declared hash
    declared = hash_path.read_text().strip()
    
    # Read MRCC hash
    with open(mrcc_path) as f:
        mrcc = json.load(f)["sha256"]
    
    # Verify all match
    if computed != declared or computed != mrcc:
        raise ValueError(f"Hash mismatch: computed={computed}, declared={declared}, mrcc={mrcc}")
    
    return True

Shell

#!/bin/bash
set -e

COMPUTED=$(sha256sum artifacts/base120.v1.0.0.seed.json | cut -d' ' -f1)
DECLARED=$(cat artifacts/base120.v1.0.0.seed.sha256)
MRCC=$(jq -r '.sha256' compliance/base120.v1.0.0.seed.mrcc.json)

if [ "$COMPUTED" != "$DECLARED" ] || [ "$COMPUTED" != "$MRCC" ]; then
    echo "FAIL: Hash mismatch"
    exit 1
fi

echo "PASS: Seed integrity verified"

Consumption Patterns

Clone or fetch the seed directly from this repository:

git clone https://github.com/hummbl-dev/base120.git
# Verify integrity before use

Pattern 2: Cached Copy

Store a local copy with provenance:

import json
from datetime import datetime, timezone

# After verification
provenance = {
    "source": "hummbl-dev/base120",
    "version": "v1.0.0",
    "sha256": "74c51092b218dcf7b430569fffb36a23ae42aa07f7f1b900479b1721e585656d",
    "acquired_at": datetime.now(timezone.utc).isoformat(),
    "verified": True
}

Pattern 3: API Fetch

import requests

SEED_URL = "https://raw.githubusercontent.com/hummbl-dev/base120/main/artifacts/base120.v1.0.0.seed.json"
HASH_URL = "https://raw.githubusercontent.com/hummbl-dev/base120/main/artifacts/base120.v1.0.0.seed.sha256"

# Fetch and verify before use

Model Access

The seed contains 120 models across 6 domains:

Domain Prefix Count Description
Perspective P 20 P1-P20
Inversion IN 20 IN1-IN20
Composition CO 20 CO1-CO20
Decomposition DE 20 DE1-DE20
Reasoning RE 20 RE1-RE20
Systems SY 20 SY1-SY20

Lookup by ID

import json

with open("artifacts/base120.v1.0.0.seed.json") as f:
    corpus = json.load(f)

# Build lookup index
models_by_id = {m["id"]: m for m in corpus["models"]}

# Access specific model
model = models_by_id["P1"]
print(model["name"])  # "First Principles Framing"
print(model["definition"])

Lookup by Domain

from collections import defaultdict

models_by_domain = defaultdict(list)
for m in corpus["models"]:
    models_by_domain[m["domain"]].append(m)

# Get all Composition models
composition_models = models_by_domain["CO"]

Citation Format

When referencing Base120 models in downstream systems, use:

base120:v1.0.0:{model_id}

Examples:

Prohibited Actions

Downstream systems MUST NOT:

Prohibition Rationale
Modify the seed JSON Immutable trust anchor
Add or remove models Governance violation
Rename or alias model IDs Shadow vocabulary
Extend model definitions Semantic purity
Provide feedback to Base120 One-way consumption
Cache with TTL refresh v1.0.0 is frozen

Integration Template

For comprehensive integration guidance, see the 90-Day Downstream Integration Template in System B.

Support

Base120 v1.0.0 is frozen and governance-complete. For questions:

This site is open source. Improve this page.