REFERENCE ARCHITECTURE AUDITS
Architecture Audits & Scorecards
Dogfood and reference audit scorecards run across production codebases (COMPASS, Elysium, Card Vault) and early partner suites. Run npx chemx audit --share to test your repository and publish results.
Reference & Community Reports
Author dogfood test runs and shared scorecards ingested from Chemical-X-Protocol GitHub Discussions.
showcase#5
Chemical-X-Protocol/starter-kit
MHI Score:100 / 100 [A+]
5,531 LOC 37 files 2% (53,741 tokens)
@xopherdeepteardown#6
HalloftheGods/elysium
MHI Score:0 / 100 [F]
142,233 LOC 826 files 7% (1,173,683 tokens)
@xopherdeeptransformation#4
Chemical-X-Protocol/starter-kit
MHI Score:80 -> 100 [B -> A+]
5,531 LOC 37 files +20 MHI (-2,861 tokens)
@xopherdeepshowcase#3
Chemical-X-Protocol/starter-kit
MHI Score:99 / 100 [A+]
5,400 LOC 36 files Zero High/Critical
@xopherdeepteardown#2
Chemical-X-Protocol/starter-kit
MHI Score:65 / 100 [D]
5,200 LOC 35 files 15 Hazards
@xopherdeep ZERO-INSTALL QUICKSTART RUN IN 2 SECONDS
Test It on Your Worst Repo
Zero configuration. Zero project dependencies. Run npx chemx audit directly in any project root to inspect token bloat, AI slop, and component blast radius.
$npx chemx audit
██████╗██╗ ██╗███████╗███╗ ███╗██╗ ██████╗ █████╗ ██╗ ██╗ ██╗
██╔════╝██║ ██║██╔════╝████╗ ████║██║██╔════╝██╔══██╗██║ ╚██╗██╔╝
██║ ███████║█████╗ ██╔████╔██║██║██║ ███████║██║ ╚███╔╝
╚██████╗██║ ██║███████╗██║ ╚═╝ ██║██║╚██████╗██║ ██║███████╗ ██╔╝ ██╗
MOLECULAR HEALTH INDEX: 94 / 100 [Grade: A • Highly Modular]
Source Files Analyzed: 54 files (4,120 LOC)
Token Reduction Potential: 36.8% (Estimated 48,000 excess tokens saved)
✔ PASSED: Line Budgets & Capsule Boundaries (< 100 LOC)
✔ PASSED: Two-Stage Atomic Booleans & Result Tuples
✔ PASSED: AI Slop & Conversational Residue Check
✔ PASSED: Encapsulated Primitive Auditing EMPIRICAL CONTEXT BENCHMARK KILLER METRIC
17,000 Tokens → 45 Tokens
When an agent runs npm test, it wastes half its context window reading green checkmarks. chemx verify strips the noise and returns surgical, machine-readable diagnostics only when something breaks.
| Verification Action | Raw Command Overhead | Chemical X (chemx) | Context Saved |
|---|---|---|---|
Test Runner (143 specs) chemx test | 957 lines (~12,960 tokens) via npm test / vitest | 1 summary line (~12 tokens) silent exit / surgical failure | 99.9% token reduction |
TypeScript Typecheck chemx typecheck | Multi-line banner spam via tsc --noEmit | Silent exit / JSON diagnostics (~10 tokens) silent exit / surgical failure | Direct agent ingestion |
Full Verify Pipeline chemx verify | ~17,000+ tokens burned via audit + tsc + test | ~45 tokens total silent exit / surgical failure | 99.7% context preserved |
Automated Token Telemetry & CI/CD PR Cost Accounting
Runs in pre-commit hooks, GitHub Actions, and agent terminals