etyb-skills
Health Warn
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 8 GitHub stars
Code Pass
- Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Pass
- Permissions — No dangerous permissions requested
This tool provides a collection of 30 behavioral skills, engineering disciplines, and automated hooks for AI coding agents (like Claude Code and OpenAI Codex). It acts as a virtual engineering team to enforce structured development workflows, such as requiring tests before code and reviews before commits.
Security Assessment
Overall risk: Low. The automated code scan of 12 files found no dangerous patterns, no hardcoded secrets, and the tool does not request any dangerous permissions. However, installing the framework involves executing remote shell scripts (`install.sh`, `install-codex-runtime.sh`). Users should always review these bash scripts before running them on their local machine or in a development environment.
Quality Assessment
The project is actively maintained, with its most recent push occurring today. It is properly licensed under the permissive and standard MIT license. The primary concern is its extremely low community visibility; it currently has only 8 GitHub stars. Because it is a new and small project, it has not yet been widely vetted by the broader developer community.
Verdict
Safe to use, but you should review the installation shell scripts before executing them, and keep in mind that the framework has minimal community validation.
ETYB — virtual engineering company for AI coding agents. 31 skills: CTO + 20 specialists + 9 engineering disciplines. Claude Code, Codex, Antigravity.
etyb.ai · v2.1.0 · Changelog · Install Guide · Architecture
etyb-skills
Install a virtual engineering company. Your AI agent gets a CTO, 20 specialists, and 9 always-on engineering disciplines — out of the box.
30 coordinated skills that turn any agentskills.io-compliant agent into a team that ships features methodically: brainstorming before architecture, tests before code, reviews before commits, evidence before "done."
Quick Start
# Claude Code — native plugin
/plugin marketplace add e-t-y-b/etyb-skills
/plugin install etyb-full@etyb-skills
# OpenAI Codex, Google Antigravity, or manual install
git clone https://github.com/e-t-y-b/etyb-skills.git
./etyb-skills/scripts/install.sh
# Codex projects — add runtime hooks + custom agents
./etyb-skills/scripts/install-codex-runtime.sh --target /path/to/your-project
Platform Support
| Platform | Enforcement | What You Get |
|---|---|---|
| Claude Code | Hook-enforced (flagship) | Deterministic gates via PreToolUse/PostToolUse hooks — edit-before-test, pre-merge, pre-commit review checks |
| OpenAI Codex | Partial runtime-enforced | 4 lifecycle hooks (prompt guardrails, Bash guards, stop checks) + 4 custom agents + per-skill openai.yaml metadata. Documented model-trusted gaps |
| Google Antigravity | Model-trusted | Markdown-first protocols; ADK integration deferred. All gates and disciplines apply via instruction |
Who This Is For
- Teams shipping real software — your AI agent works like a disciplined team member, not a solo cowboy
- Engineers tired of AI "yes-and" behavior — agents that skip tests, rubber-stamp reviews, or chase symptoms
- Regulated or high-stakes codebases (fintech, healthcare, e-commerce) — traceable decisions, gated releases, evidence-backed claims
- Solo developers — every specialist works standalone; the CTO is optional
What You Get
An AI coding agent that works like a 100-person engineering org:
- Refuses to ship untested code — TDD enforcement with deterministic hooks, not just instructions
- Stops you from building the wrong thing — structured brainstorming before architecture
- Pushes back on bad review feedback — evaluates findings on merit, no performative agreement
- Coordinates parallel work — subagent dispatch with two-stage review and worktree isolation
- Covers the full SDLC — from research through production operations
- Knows your domain — fintech ledgers, HIPAA compliance, e-commerce patterns, real-time systems
Architecture
USER REQUEST
|
ETYB (CTO — routes, enforces gates, tracks plans)
|
| ALWAYS-ON PROTOCOL LAYER
| |-- TDD — no code without failing test
| |-- Verification — evidence before claims
| |-- Review — no performative agreement
| |-- Plan execution — one task at a time
| |-- Brainstorm-first — explore before solving
| |-- Branch safety — never merge without green tests
| |-- Subagent coordination — parallel dispatch + review
| |-- Self-improvement — failing eval before skill changes
| +-- Debugging — root-cause-first after repeated failures
|
DOMAIN EXPERTS (14 core teams + 6 vertical specialists)
|
DEEP REFERENCES (100+ files, loaded on demand)
The 30 Skills
ETYB (1)
Your virtual CTO. Routes requests, enforces 5-phase gates (Design > Plan > Implement > Verify > Ship), mandates experts, tracks living plans. Works standalone, or as the conductor that pulls the other 29 skills into a coordinated team.
Core Team (14)
| Skill | Phase | What It Does |
|---|---|---|
research-analyst |
Discovery | Tech evaluation, competitive analysis, feasibility, requirements |
project-planner |
Planning | Sprint planning, timelines, agile coaching |
system-architect |
Design | System design, domain modeling, API design, data architecture |
frontend-architect |
Design + Dev | React, Angular, Vue, Svelte, SEO, performance, accessibility |
backend-architect |
Design + Dev | Java, TypeScript, Go, Python, Rust, microservices, auth |
database-architect |
Design + Dev | SQL, NoSQL, caching, search, data pipelines, migrations |
mobile-architect |
Design + Dev | React Native, Flutter, iOS, Android, mobile performance |
ai-ml-engineer |
Design + Dev | ML, MLOps, LLMs, data science, AI product integration |
qa-engineer |
Testing | Unit, integration, E2E, performance, API testing, test strategy |
devops-engineer |
Deploy | CI/CD, containers, Kubernetes, AWS/GCP/Azure, IaC, releases |
sre-engineer |
Operations | Monitoring, logging, tracing, incident response, chaos engineering |
security-engineer |
Cross-cutting | AppSec, infra security, IAM, compliance, threat modeling |
technical-writer |
Cross-cutting | API docs, architecture docs, runbooks, user guides |
code-reviewer |
Cross-cutting | Code quality, performance, security, architecture review |
Domain Specialists (6)
| Skill | Domain |
|---|---|
social-platform-architect |
Feeds, social graphs, content ranking, fan-out, real-time delivery |
e-commerce-architect |
Catalogs, cart/checkout, payments, inventory, order management |
fintech-architect |
Ledgers, payment processing, AML/KYC, PCI/PSD2, fraud detection |
saas-architect |
Multi-tenancy, billing, subscriptions, onboarding, usage metering |
real-time-architect |
WebSockets, CRDTs, collaboration, gaming backends, live streaming |
healthcare-architect |
HIPAA, HL7/FHIR, EHR integration, patient data, audit trails |
Process Protocols (9)
| Skill | Always On | Runtime Support |
|---|---|---|
tdd-protocol |
Every code change | Claude hooks, Codex prompt/Bash guardrails, Antigravity model-trusted |
review-protocol |
Every review cycle | Claude pre-commit hook, Codex reviewer agent + commit reminder, Antigravity model-trusted |
subagent-protocol |
Parallel work | Claude isolated subagents, Codex custom agents, Antigravity markdown-first |
git-workflow-protocol |
Branch management | Claude pre-merge hook, Codex merge guard via Bash hooks, Antigravity model-trusted |
plan-execution-protocol |
Active plans | Claude native plan mode + post-edit hook, Codex .etyb/plans/, Antigravity .etyb/plans/ |
brainstorm-protocol |
Ambiguous requests | Platform-neutral |
skill-evolution-protocol |
Skill improvements | Platform-neutral |
verification-protocol |
Every completion claim | Claude deterministic, Codex stop hook assist, Antigravity model-trusted |
debugging-protocol |
Active troubleshooting | Platform-neutral |
Install
Claude Code (plugin)
# Everything
/plugin marketplace add e-t-y-b/etyb-skills
/plugin install etyb-full@etyb-skills
# Or install subsets
/plugin install etyb-process-protocols@etyb-skills # 9 protocols + ETYB
/plugin install etyb-core-team@etyb-skills # 14 core teams + ETYB
/plugin install etyb-verticals@etyb-skills # 6 domain specialists
Codex, Antigravity, or Manual
git clone https://github.com/e-t-y-b/etyb-skills.git
cd etyb-skills
./scripts/install.sh # detects platform, installs skills
./scripts/install.sh --dry-run # preview without writing
Codex Runtime (hooks + agents)
./scripts/install-codex-runtime.sh --target /path/to/your-project
Installs .codex/config.toml, lifecycle hooks, and 4 custom agents (explorer, planner, reviewer, docs researcher). Backs up existing .codex/ on conflict. See docs/installation.md for the full guide.
Updating
./scripts/update.sh --check # is there a newer version?
./scripts/update.sh # interactive update (shows before/after)
./scripts/update.sh --force # skip confirmation prompts
Preserves .etyb/plans/, .claude/plans/, and .claude/settings.local.json. Uses git merge --ff-only — no destructive operations. See CHANGELOG.md for version history.
How Skills Load (Token Efficiency)
Skills use progressive disclosure — a markdown-based RAG pattern:
Layer 0 Runtime guardrails (0 tokens — scripts outside the LLM)
Layer 1 ETYB always loaded (~3,500 tokens — the culture)
Layer 2 Relevant skill loads on demand (~2,500 tokens — the router)
Layer 3 Single reference loads on demand (~4,000 tokens — deep knowledge)
Per-activation: ~6,000-10,000 tokens (not the whole system)
Tier 0-1 requests: 0 extra tokens (ETYB handles directly)
Evidence: With vs Without Skills
| Scenario | Without Skills | With Skills |
|---|---|---|
| "Skip tests, demo tomorrow" | Wrote production code immediately | Refused. Named the rationalization. Laid out 4 TDD cycles. |
| "Build restaurant app with React+Firebase" (6 features) | Built entire architecture, 6-week plan, all code | Stopped. Identified 3 red flags. Asked 10 questions. Challenged build-vs-buy. |
| "Handle these 5 review findings" | Agreed with all 5 | Pushed back on 2 with evidence. Caught a mis-severity. Demanded a test before accepting a code change. |
Average token overhead: +53%. What you get: an engineer who says "no" when it matters.
Links
- Website: etyb.ai
- Repo: github.com/e-t-y-b/etyb-skills
- Latest release: v2.1.0
- Changelog: CHANGELOG.md
- Install guide: docs/installation.md
- Architecture: docs/architecture.md
- Issues & contributions: github.com/e-t-y-b/etyb-skills/issues
License
MIT — see LICENSE.
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found