Everything Gemini

---

⚡ What is Everything Gemini?

Google Antigravity is Google’s state-of-the-art AI-first coding platform, powered by Gemini models, an event-driven execution loop (Cortex), progressive disclosure, multi-surface orchestration (Desktop 2.0, VS Code-based IDE, agy CLI, and Python SDK), and localized Knowledge Items (KI).

Out of the box, Antigravity is a formidable pair programmer. But without structured harnesses, agents can suffer from:

• Context bloat: Loading massive guidelines into every prompt turn, eroding the model's reasoning window.
• Unchecked execution risks: Accidentally running destructive terminal commands or pushing unfinished code.
• Premature stops: Quitting multi-step tasks before tests pass or edge cases are verified.
• Amnesia between sessions: Forgetting architectural lessons learned yesterday.

Everything Gemini (everything-gemini) solves this. It provides a battle-tested harness system consisting of:

1. 🛡️ Safety Shield & Lifecycle Hooks (hooks.json): Deterministic gating for destructive commands, auto-lint formatters, and stop-guard verifiers.
2. 🧠 Progressive Disclosure Skills (skills/): On-demand procedural playbooks that load only when needed, saving up to 85% in token overhead.
3. 📏 Hierarchical Rules (GEMINI.md / AGENTS.md): Context-aware coding standards, zero-hallucination protocols, and clean-architecture rules.
4. 🔌 Plug-and-Play Plugins (plugins/): Domain bundles for Fullstack Web, Python AI, DevOps, and Security.
5. 🌐 Curated MCP Servers (mcp/): Production blueprints for Persistent Semantic Memory (@modelcontextprotocol/server-memory), High-Performance Cache (@modelcontextprotocol/server-redis), Container Engine (@modelcontextprotocol/server-docker), Real-Time OLAP (@modelcontextprotocol/server-clickhouse), Web Fetch (server-fetch), GitHub, PostgreSQL, and SQLite.
6. 🤖 Subagent & Browser Automation: Specialized recipes for autonomous browser subagents with automated WebP recording.

---

🚀 Quickstart

1. Instant Installation via npx (Zero Setup)

Run this directly in any project directory or terminal:

npx everything-gemini install

2. One-Line Interactive Installer

# macOS / Linux
curl -fsSL https://raw.githubusercontent.com/salmandev/everything-gemini/main/scripts/install.sh | bash

# Windows (PowerShell)
irm https://raw.githubusercontent.com/salmandev/everything-gemini/main/scripts/install.ps1 | iex

3. Via Repository Clone

# Clone the repository
git clone https://github.com/salmandev/everything-gemini.git
cd everything-gemini

# Run the Everything Gemini CLI
./bin/everything-gemini install --profile developer

🛡️ Instant PHP2Shell Audit (CVE-2026-109618 / CVE-2026-109619)

Perform non-intrusive defensive security audits on public endpoints and private servers without installing anything:

# Audit any public website or API endpoint
npx everything-gemini phpcheck "https://example.com"

# Audit local server, container, and detect un-restarted PHP-FPM workers in RAM
npx everything-gemini phpcheck --server

Read the full PHP2Shell Security Advisory & Social Media Launch Kit for distribution backport matrices and ready-to-post copy.

4. Installation Profiles

Profile	Target Audience	What Gets Installed
core	Minimalists & Speed	Safety-shield hook, TDD skill, repo-onboarding skill, baseline GEMINI.md.
developer (Recommended)	Daily Software Engineers	Core + 28 Scoped Subagents (agents/), Autonomous Company OS & C-Suite Avatars (ceo, cto, ciso, cpo, fullstack-developer, automation-tester), Frontier Skills (Agency Boardroom, Sprint Runner, Gemini Game Studios, Gemini 4 Argon 3D & Spatial, Google Elephant-Goldfish Model, Superpowers TDD methodology, Darwin prompt evolution, CVE mitigator, Multimodal vision, 2M context caching, Subagent handoff, Security audit), Fullstack, 3D Games & Python AI plugins, Auto-formatter & Goal-verifier hooks.
full	Lead Engineers & Archs	Everything: all 117 progressive skills, 28 scoped subagents, full lifecycle hooks suite (13 hooks), 14 MCP blueprints, all plugins including Google Cloud, Agent Platform, and Gemini 4 Argon 3D Suite, configs, and templates (The 158-Component Landmark Release).

5. In CI / GitHub Actions Workflows

Automate Everything Gemini skill, subagent, and lifecycle safety hook provisioning directly in your CI pipelines using our official composite action:

- name: Setup Everything Gemini
  uses: salmandev/everything-gemini@v1
  with:
    profile: developer # core | developer | full
    global: false      # true for ~/.gemini/config/, false for active workspace .agents/

---

🏗️ Architecture & Internals

everything-gemini is engineered specifically around Google Antigravity's internal execution engine:

flowchart TB
    subgraph AntigravitySurfaces["1. Antigravity Surfaces"]
        Desktop["Antigravity 2.0 (Desktop)"]
        IDE["Antigravity IDE (VS Code Fork)"]
        CLI["Antigravity CLI ('agy')"]
        SDK["Python SDK ('google-antigravity')"]
    end

    subgraph CustomizationRoots["2. Customization Roots"]
        Global["~/.gemini/config/ (Global)"]
        Workspace[".agents/ (Workspace Root)"]
    end

    subgraph GeminiHarness["3. Everything Gemini Harness & Engine"]
        direction TB
        HookEngine["Lifecycle Hooks Engine\n(PreToolUse | PostToolUse | PreInvocation | Stop)"]
        Progressive["Progressive Disclosure Router\n(Injects only metadata until model decides to view)"]
        MemoryBridge["Memory Bridge\n(Knowledge Items [KI] + Brain Artifacts + Transcripts)"]
    end

    subgraph Capabilities["4. Agent Capabilities & Sandboxes"]
        Subagents["Autonomous Subagents\n(browser_subagent, test-runner)"]
        MCP["Model Context Protocol (MCP)\n(Stdio & SSE Servers)"]
        TerminalSandbox["Terminal Sandbox & Execution Policy"]
    end

    AntigravitySurfaces --> CustomizationRoots
    CustomizationRoots --> GeminiHarness
    GeminiHarness --> Capabilities

Key Differences: ECC vs. Everything Gemini

📖 Read the Comprehensive Architectural Breakdown: Everything Claude Code (ECC) vs. Everything Gemini

Feature	Everything Claude Code (ECC)	Everything Gemini (everything-gemini)
Target Platform	Claude Code CLI & harnesses	Antigravity 2.0 Desktop, IDE, agy CLI, Python SDK
Model Engine	Claude 3.5 / 3.7 Sonnet	Gemini 3.8 / 1.5 Flash & Pro
Context Protocol	Direct markdown prompts	Progressive Disclosure (Metadata injected first, bodies lazy-loaded)
Context Window & Caching	200k tokens, ephemeral	1M-2M+ tokens with Native Prefix Context Caching (~75% cost savings)
Memory System	Flat markdown files	Knowledge Items (KI) + Adaptive Learned Instincts
UI & Visual Multimodality	Standard text output	Native browser_subagent + WebP Screen Video Recording + Vision
Subagents Fleet	Manual sub-prompts	28 Scoped Subagents (agents/) with Formal Handoff Protocol & Autonomous C-Suite Avatars
Lifecycle Hooks	Shell scripts	Typed protojson contracts (PreToolUse, PostToolUse, PreInvocation, PostInvocation, Stop)
Tool Integration	Claude tool use	MCP standard (Stdio & SSE) with auto-discovery & namespacing

---

📂 Repository Directory Layout

everything-gemini/
├── .github/                      # CI/CD workflows & issue templates
├── bin/                          # Zero-dependency harness & scaffolding CLI
├── agents/                       # Scoped subagent fleet (Planner, Reviewer, Release Engineer, etc.)
├── scripts/                      # Automated installers and harness validators
├── plugins/                      # Reusable modular plugins (web-fullstack, python-ai, security-shield, google-cloud)
├── skills/                       # 117 Progressive disclosure skills (108 core + 9 plugins)
├── rules/                        # Contextual guidelines & standards
├── hooks/                        # Event-driven lifecycle hooks (GeminiShield, Auto-lint, ThinkingBudget)
├── mcp/                          # Production-ready MCP server blueprints
├── docs/                         # The Comprehensive Antigravity Bible
└── templates/                    # Starter templates (minimal, recommended, full-power)

---

🧰 Customization Catalog

🤖 Scoped Sub-Agents (agents/)

Everything Gemini provides specialized subagent specifications designed for Antigravity's native invoke_subagent delegation and Python SDK orchestration:

Subagent	Role	Focus Area
planner	Strategic Planner	Dependency mapping, impact radius analysis, phased checklists.
code-reviewer	Quality Auditor	Edge case verification, type safety, regression prevention, minimal diffs.
build-repair	Build Triage Specialist	Compiler error diagnosis, test failure auto-repair, surgical fixes.
browser-qa	Automated Visual QA	Native browser testing with automated WebP recording into artifacts.
security-auditor	Security Auditor	Credential scanning, OWASP Top 10, dependency audit, permission gates.
architect	Systems Architect	C4 Mermaid blueprints, sequence diagrams, and formal ADR records.
database-migrator	Database Schema Specialist	Zero-downtime expand/contract SQL migrations, lock analysis, and rollbacks.
performance-tuner	Latency & Profiling Expert	CPU/memory flamegraphs, database execution plans, and N+1 waterfalls.
documentation-writer	Technical Documentation Specialist	API specifications, ADR governance, changelogs, and clickable file:// link fidelity.
release-engineer	Release Automation Gatekeeper	Semantic Versioning, build artifact checksums, git release tagging, and canary rollouts.
incident-responder	SRE & Incident Commander	Alert triage, OTel trace correlation, blast radius containment, blameless postmortems.
platform-engineer	Cloud Platform & IDP Specialist	Terraform/OpenTofu IaC, Kubernetes manifests, zero-trust network policies, container right-sizing.
mlops-engineer	MLOps & LLMOps Specialist	Model serving (Triton/vLLM), vector clustering (ChromaDB/Milvus), prefix context caching, KEDA autoscaling.
swarm-coordinator	Swarm Intelligence Coordinator	Ephemeral worker synthesis, parallel worktree dispatch, Byzantine consensus voting, and zero-leak reclamation.
resilience-engineer	Chaos & Resilience Specialist	Harness mutation testing, adversarial fault injection, circuit breaker failover auditing, and zero-downtime resilience certification.
cross-repo-orchestrator	Cross-Repository Swarm Architect	Multi-repository dependency synchronization, cross-boundary 2PC commit gating, and contract schema alignment.
darwin-prompt-engineer	Darwinian Genetic Prompt Architect	Multi-generation prompt tournament selection, adversarial fuzzing, and Pareto efficiency optimization.
cve-mitigator	Zero-Day CVE Remediation Specialist	4-layer defense-in-depth virtual patching, kernel ASLR verification, container least-privilege, and memory inode auditing.
elephant-architect	Deep-Context Technical Architect	Long-horizon design exploration, requirements decomposition, living design documents under the strict "NO CODE" invariant.
goldfish-reviewer	Zero-Memory Adversarial Reviewer	Stateless, unbiased peer review across 4 lenses (Design Critic, Naive Reader, Tech Lead, Codebase Auditor) eliminating sycophancy.
spatial-game-architect	3D Graphics & Game Architect	Three.js, WebGPU, Rapier/Cannon physics, GLSL/TSL shaders, procedural mesh synthesis, and multi-turn visual critique loops with Gemini 4 Argon.
game-director	Executive Game Director & Studio Producer	Orchestrates multi-department game development, GDD authorship, core loops, milestone roadmaps, and 60 FPS gold release certification.
ceo	Chief Executive Officer	Strategic vision, OKRs, MVP feature scoping, executive consensus, and final ship/no-ship authority.
cto	Chief Technology Officer	Architecture blueprints, ADR governance, tech debt containment, and non-functional requirements.
ciso	Chief Information Security Officer	Zero-trust defense, STRIDE threat modeling, OWASP compliance, and unilateral security veto.
cpo	Chief Product Officer	PRD authorship, user journey mapping, feature prioritization, and BDD acceptance criteria.
fullstack-developer	Lead Fullstack Engineer	Clean architecture, surgical diffs, backend APIs, modern frontend, and database schema integration.
automation-tester	Senior QA Automation Engineer	Playwright E2E suites, visual regression pixel testing, boundary fuzzing, and release certification.

1. Skills (skills/)

Skills in Antigravity are lazy-loaded runbooks. Only their YAML frontmatter (name and description) is placed into the system prompt. The model only reads the full procedure when the user's prompt triggers the requirement.

• elephant-goldfish-model: Implements Google's Elephant-Goldfish Model (Dave Rensin) for spec-driven development. Couples long-horizon deep-context Elephant architecture sessions (strict NO CODE rule) with stateless zero-memory Goldfish adversarial review swarms across 4 distinct critical lenses.
• superpowers: Enforces disciplined agentic software development methodology (inspired by Jesse Vincent's obra/superpowers). Conducts interactive specification gathering, creates step-by-step red-green TDD implementation plans, orchestrates scoped subagents, and ensures zero-drift verification before committing changes.
• autonomous-incident-healer: Autonomously detects production incident anomalies, executes deterministic self-healing runbooks, triggers automated canary traffic draining or rollback, and compiles blameless root-cause analysis (RCA).
• spanner-distributed-locks: Implements enterprise distributed locks, leader election, and mutual exclusion leases on Google Cloud Spanner using TrueTime and conditional mutations.
• gemini-live-audio-transcriber: Streams bidirectional low-latency audio via WebSocket/WebRTC to the Gemini Live API, handling Opus framing, Voice Activity Detection (VAD), live turn interruptions, and real-time pair-programming transcription.
• multi-modal-screen-actor: Executes autonomous GUI perception and mouse/keyboard action loops for Desktop 2.0, normalizing screen coordinates (0-1000), grounding UI elements, and verifying visual side-effects.
• spanner-vector-search: Executes high-performance Approximate Nearest Neighbor (ANN) vector search and hybrid Graph RAG queries on Google Cloud Spanner using ScaNN indexes and COSINE_DISTANCE functions.
• cloud-deploy-canary: Orchestrates automated progressive delivery, canary traffic splitting, and metric-based rollback gates using Google Cloud Deploy for Cloud Run and GKE.
• alloydb-omni-vector: Configures AlloyDB Omni with pgvector and the Columnar Engine for ultra-fast HNSW vector similarity search and analytical queries.
• zero-trust-workload-identity: Configures GCP Workload Identity Federation, SPIFFE/SPIRE cryptographic attestation, and short-lived STS tokens for multi-cloud autonomous agents without static service account keys.
• cve-remediation-orchestrator: Defense-in-depth CVE mitigation, virtual patching, kernel ASLR hardening, and worker memory verification.
• php2shell-version-auditor: Enterprise vulnerability auditor and fleet triage runbook for PHP2Shell (CVE-2026-109618 & CVE-2026-109619).
• phpcheck-url: Fast non-intrusive remote HTTP banner and advertised PHP version auditor.
• phpcheck-server: Server-side OS package backport auditor and un-restarted worker memory detector.
• adversarial-fuzz-generator: Synthesizes property-based and adversarial test suites using hypothesis-style fuzzing to mathematically verify codebase and prompt modifications.
• genetic-prompt-cross-pollinator: Cross-pollinates winning agent prompt traits and subagent contracts using genetic crossover operators and Pareto selection.
• instinct-distillation-compiler: Compiles runtime edge cases and failure modes into persistent, deterministic learned rules in .agents/hooks/learned-instincts/.
• gemini-live-screen-vision: Streams active IDE screens and browser DOM states to Gemini Live, analyzing real-time repaints, terminal error bursts, and layout regressions.
• cross-repo-swarm-sync: Coordinates distributed multi-repository agent swarm synchronization, atomic dependency alignment, and cross-boundary git branch PR orchestration via Google Cloud Pub/Sub and Cloud Run.
• ebpf-microvm-sandbox: Enforces rootless ephemeral Micro-VM isolation (Firecracker/gVisor), Linux Landlock filesystem jails, and Cilium eBPF network filtering for zero-trust speculative tool and code execution.
• context-distillation-compressor: Performs AST-guided semantic context distillation, lossy-to-lossless prompt compaction, and memory pruning to prevent context bloat and token degradation in long agent trajectories.
• speculative-decoding-orchestrator: Orchestrates dual-model speculative decoding routing draft tokens via Gemini 2.0 Flash and parallel verification via Gemini 3.8 Extended Thinking for 3.5x inference acceleration.
• webrtc-agent-gateway: Real-time bidirectional WebRTC audio duplex gateways, Opus audio transcoding, STUN/TURN ICE candidate negotiation, and sub-200ms voice streaming for Gemini Live pair programming.
• opentelemetry-agent-observability: OpenTelemetry GenAI semantic conventions, distributed multi-agent W3C trace propagation, tool call span attributes, and Google Cloud Trace export.
• harness3-mutation-testing: Autonomous mutation testing, adversarial fault injection, schema perturbation, and mathematical resilience scoring across agent prompts and tool surfaces.
• envoy-ai-gateway: Envoy Proxy AI Gateway for Gemini models with token-aware rate limiting, multi-model circuit breakers, active health checks, and zero-trust mTLS termination.
• ephemeral-swarm-orchestrator: Dynamically synthesizes, deploys, and reclaims ephemeral multi-agent swarms for cross-repository parallel refactoring and consensus aggregation.
• dynamic-thinking-allocation: Profiles cognitive task complexity and dynamically routes between low-latency Gemini Flash and Gemini 3.8 Extended Thinking reasoning budgets.
• ebpf-sandbox-enforcer: Enforces zero-trust rootless micro-VM and eBPF kernel security boundaries for speculative agent tool executions and container sandboxes.
• quantum-safe-cryptography: Implements NIST FIPS 203/204 Post-Quantum Cryptography (PQC) algorithms, hybrid TLS 1.3 key exchange, and quantum-resistant API authentication.
• gemini-live-multimodal: Real-time bidirectional WebSocket streaming audio and video pair programming with Gemini Live API, ephemeral client tokens, and Voice Activity Detection (VAD).
• cost-oracle-finops: Pre-flight predictive token cost modeling, spend budgeting, and pricing tier optimization across Gemini 2.0 Flash, Pro, and Gemini 3.8 extended reasoning agents.
• vertex-ai-feature-store: Provisions and queries Google Cloud Vertex AI Feature Store, point-in-time correct historical joins, low-latency online serving, and Gemini embedding feature tables.
• spanner-change-streams: Google Cloud Spanner Change Streams for real-time transaction log Change Data Capture (CDC), Pub/Sub integration, and event-driven database synchronization.
• bigquery-vector-search: Native BigQuery vector search, VECTOR_SEARCH SQL statements, ML.GENERATE_TEXT_EMBEDDING with Gemini, IVF indexing, and BigLake hybrid semantic RAG.
• cloud-run-agents: Containerized autonomous agent hosting on Google Cloud Run, Cloud Run Jobs for Harness² recursive probing, Eventarc triggers, and NVIDIA L4 GPU acceleration.
• pubsub-event-mesh: Asynchronous event mesh architectures, exactly-once delivery, dead-letter queues (DLQ), and schema-validated multi-agent message routing.
• spanner-graph-rag: Distributed Spanner Graph schemas, ISO GQL queries, knowledge graph traversal, and hybrid graph-vector RAG pipelines for Gemini agents.
• context-warmer: Automated workspace repository analysis, Knowledge Item clustering, and pre-warming prefix context caches before subagent execution.
• harness2-recursive-optimization: Google Research Harness² framework (Harness²: Recursive Agent Harnessing for an Open World, arXiv:2610.14224), 8-surface architecture, test-time Propose-Probe-Compose recursive loops, and continual harness learning.
• google-flow-orchestrator: Google Agentic Flow state machine orchestration, cyclic multi-agent graph execution, conditional state routing, and human-in-the-loop pauses.
• dataproc-spark-analytics: Google Cloud Dataproc Serverless Spark, PySpark batch pipelines, Gemini embedding UDFs, BigLake vector indexing, and large-scale probe score verification.
• keda-event-autoscaling: Kubernetes Event-driven Autoscaling (KEDA), ScaledObjects/Jobs, Kafka/Redis/Prometheus triggers, scale-to-zero.
• chromadb-vector-store: Persistent vector storage, Gemini embeddings, collection partitioning, metadata filtering, and semantic similarity search.
• opa-gatekeeper-policy: Open Policy Agent (OPA) Gatekeeper ConstraintTemplates and Rego admission policies for Kubernetes governance.
• triton-inference-server: NVIDIA Triton Inference Server multi-model serving, dynamic batching, GPU acceleration, TensorRT-LLM, and vLLM backends.
• milvus-vector-clustering: Enterprise Milvus vector clustering, HNSW/IVF-PQ indexing, hybrid dense-sparse search, and partition keys.
• cloudflare-r2-object-storage: Zero-egress fee object storage, S3-compatible APIs, pre-signed upload URLs, and Workers edge CDN.
• vault-secrets-management: HashiCorp Vault enterprise secrets orchestration, AppRole authentication, dynamic ephemeral DB credentials, and PKI mTLS.
• distributed-rate-limiting: High-throughput distributed rate limiting using Redis sliding window logs, token buckets, and RFC 6585 compliance.
• vllm-inference: High-throughput LLM serving, PagedAttention KV cache, continuous batching, and tensor parallelism.
• debezium-cdc: Change Data Capture (CDC) streaming, PostgreSQL logical decoding, Kafka Connect, and outbox patterns.
• cilium-ebpf-networking: Cloud-native eBPF networking, transparent mTLS, L7 CiliumNetworkPolicies, and Hubble observability.
• litellm-proxy-gateway: Enterprise LLM proxy gateway, multi-provider routing (Gemini, Claude, OpenAI), budget controls, and fallback cascades.
• langgraph-orchestrator: Stateful multi-agent graphs, cyclic execution loops, persistent checkpointers, and human-in-the-loop pauses.
• playwright-visual-regression: Automated visual regression testing, pixelmatch snapshot diffing, cross-browser viewports, and containerized stability.
• open-telemetry-tracing: Distributed tracing, OTLP exporters, W3C traceparent context propagation, and metric correlation.
• argo-workflows: Kubernetes-native DAG workflows, artifact passing, container steps, retry strategies, and CronWorkflows.
• gemini-multimodal: Native visual reasoning over mockups, diagrams, and automated WebP browser recordings.
• context-caching: 1M-2M token context optimization, static prefix caching, and repository indexing.
• gemini-grounding: Real-time Google Search grounding, live documentation retrieval, and fact-checking.
• rag-architect: RAG pipeline design, AST semantic chunking, hybrid vector-keyword retrieval, and Gemini re-ranking.
• gemini-function-calling: Production function declarations, structured JSON output schemas (Pydantic), and parallel tool dispatch.
• ai-first-engineering: Requirement decomposition, executable acceptance criteria (BDD), and blast-radius containment.
• agent-evals: Automated evaluation pipelines, LLM-as-a-judge scoring, golden test datasets, and prompt regression testing.
• github-ops: GitHub workflow automation, PR generation, code review triage, and semantic releases with gh CLI & GitHub MCP.
• subagent-handoff: Multi-agent task delegation protocol, scoped contexts, and report synthesis.
• api-craftsman: REST, OpenAPI 3.1 specifications, gRPC protos, and RFC 7807 error envelopes.
• database-architect: Zero-downtime migrations, concurrent indexing, and ACID schema modeling.
• turborepo-monorepo: High-performance monorepo build caching, task pipelines, package boundaries, and Docker pruning.
• dbt-transformation: Modular SQL transformations, Jinja macros, semantic layer models, schema tests, and docs.
• apache-kafka-streams: Exactly-once event streaming, stateful window aggregations, consumer group rebalancing, and Schema Registry.
• grpc-connect-rpc: Lightweight, browser-compatible gRPC and Protocol Buffers with Connect-ES / Connect-Go and HTTP/2-3 streaming.
• duckdb-local-olap: Embedded in-process SQL analytics, direct S3/Parquet querying, and zero-copy Arrow integration.
• nats-jetstream: Ultra-low latency event streaming, message deduplication, durable consumer groups, and KV buckets.
• axum-web-service: Production Rust web APIs using Axum, Tokio async runtime, Tower middlewares, and SQLx connection pools.
• clickhouse-analytics: High-throughput OLAP schemas, MergeTree engines, Kafka ingestion, and vector search.
• temporal-workflows: Resilient distributed workflows, retry policies, async activities, and saga compensations.
• trpc-react-query: End-to-end type safety, tRPC v11 routers, and TanStack React Query v5 optimistic UI.
• wasm-rust-optimizer: High-throughput WebAssembly modules compiled from Rust with zero-copy memory buffers.
• fastapi-production: Production async FastAPI architecture, Pydantic v2 schemas, lifespan handlers, and health probes.
• redis-cache-patterns: Enterprise Redis caching strategies, distributed locking with atomic Lua, and sliding-window rate limiting.
• svelte-kit-architect: Modern SvelteKit apps with Svelte 5 runes ($state, $derived, $effect), SSR/prerender, and form actions.
• fullstack-nextjs: Next.js App Router, React Server Components (RSC), hydration error diagnosis, and Server Actions.
• graphql-architect: Schema-first GraphQL APIs, DataLoader N+1 query batching, and Relay cursor pagination.
• supabase-architect: Row Level Security (RLS) policies, pgvector embeddings, and edge function workflows.
• cloudflare-workers: Serverless edge functions, Wrangler v3, D1 SQL, and KV low-latency storage.
• docker-compose-orchestrator: Multi-container local environments (Postgres + pgvector, Redis) with healthchecks.
• kubernetes-operator: Production k8s manifests, HPA autoscaling, zero-downtime rolling updates, and health probes.
• stripe-integration: Webhook signature verification, idempotent payments, and customer portal session flows.
• devops-pipeline: Multi-stage Dockerfiles and caching GitHub Actions CI/CD workflows.
• root-cause-debugger: Regression diagnosis using binary search (git bisect) and minimal reproduction harnesses.
• tdd-workflow: Red-Green-Refactor loop with automated test runners.
• browser-qa-tester: Autonomous browser UI testing with WebP capture into artifacts.
• architecture-designer: System architecture, C4 & Mermaid blueprints.
• repo-onboarding: Fast codebase indexing into Knowledge Items (KI).
• git-craftsman: Enforces conventional commits, atomic diffs, and clean PR descriptions.
• security-audit: Scans for credentials in commit history and audits dependency risks.
• php2shell-version-auditor: Enterprise vulnerability auditor and fleet triage runbook for PHP2Shell (CVE-2026-109618 & CVE-2026-109619).
• cve-remediation-orchestrator: Defense-in-depth CVE mitigation, virtual patching, kernel ASLR hardening, and worker memory verification.
• phpcheck-url: Fast non-intrusive remote HTTP banner and advertised PHP version auditor.
• phpcheck-server: Server-side OS package backport auditor and un-restarted worker memory detector.
• performance-profiler: Identifies CPU/memory bottlenecks, analyzes database queries, and critical paths.
• continuous-learning: Codifies corrections and developer feedback into enduring Knowledge Items.
• antigravity-sdk-orchestrator: Programmatic multi-agent pipelines with the official google-antigravity Python SDK and token streaming.
• responsive-design: Fluid typography, mobile-first flex/grid layouts, media queries, and cross-browser accessibility.
• google-ax: Cluster orchestrator for Google AX (Agent Executor) on Agent Substrate, declarative ax.io/v1alpha1 manifests (Task, Workspace, Gateway, Model), sub-second suspension/resume, and zero-trust egress gating.
• agent-platform-eval-flywheel: Continuous evaluation, Golden Dataset curation, metric tracking (faithfulness, relevance), and regression detection on Vertex AI Agent Platform.
• agent-platform-rag-engine: Enterprise RAG using Vertex AI RagCorpus, hybrid vector/lexical retrieval, and grounded Gemini responses.
• developing-genkit: Production flows, type-safe tools, and multimodal agents with Google Genkit across JS/TS, Python, Go, and Dart.
• gke-ai-troubleshooting-tpu: TPU v4/v5e multi-slice hang analysis, MXLA deadlock detection, and GKE workload remediation.
• google-cloud-storage-fuse: Cloud Storage FUSE POSIX mounting, SSD caching, and high-throughput model weight loading.
• argon-3d-scene-architect: High-performance Three.js, WebGL2, and WebGPU 3D scene graph design, PBR materials, soft shadows, bloom/SSAO, and memory lifecycle disposal.
• argon-3d-game-engine: Playable real-time browser 3D games, decoupled fixed-timestep game loops, Rapier/Cannon physics, player controllers, and game state machines.
• argon-procedural-asset-generator: Algorithmic mesh generation, procedural noise heightmaps, canvas textures, and massive InstancedMesh batching without external downloads.
• argon-visual-critique-loop: Multi-turn visual perceptual inspection, camera/lighting/shadow evaluation, and autonomous iterative 3D scene refinement.
• argon-webgpu-pbr: Cutting-edge WebGPU rendering pipelines and Three.js Shading Language (TSL) node materials.
• game-studio: Virtual game development studio orchestrator, standardized GDD authoring, core loops, sprint planning, and multi-department review.
• game-audio-synthesizer: Procedural Web Audio API sound effects (SFX) and dynamic ambient background music with zero external audio assets.
• agency-boardroom: Executive C-Suite boardroom debate & consensus protocol (inspired by agency-agents). Convenes CEO, CTO, CISO, and CPO avatars to resolve strategic trade-offs, enforce security/architecture vetoes, and synthesize formal Sprint Directives.
• agency-sprint-runner: Four-stage autonomous software delivery pipeline (Spec $\rightarrow$ Build $\rightarrow$ Verify $\rightarrow$ Certify). Dispatches CPO specs to Fullstack Developer, triggers Automation QA suites, and verifies C-Suite sign-off gates.

2. Lifecycle Hooks (hooks/)

Hooks run shell scripts at critical lifecycle events with typed JSON contracts on stdin/stdout:

• gemini-shield (PreToolUse): Deterministic safety firewall intercepting run_command and denying catastrophic operations like rm -rf /, mkfs, or git push --force origin main.
• safety-shield (PreToolUse): Intercepts run_command and inspects CommandLine against strict security guidelines before shell execution.
• secret-scrubber (PreToolUse): Intercepts file write operations to block hardcoded API keys and private credentials before they reach disk.
• auto-lint-formatter (PostToolUse): Triggered after file modifications. Runs your project's linter/formatter (prettier, black, ruff, gofmt) automatically so code is always clean.
• goal-verifier (Stop): Intercepts the agent's attempt to finish. If test suites are failing or background tasks are unfinished, it returns {"decision": "continue", "reason": "Tests are still failing. Fix them before stopping."}.
• ephemeral-context (PreInvocation): Injects transient warnings or guidelines before each model reasoning step without cluttering persistent conversation history.
• learned-instincts: Session learning hook that records developer corrections and feedback into persistent instincts.
• budget-guard (PreToolUse): Intercepts tool execution to enforce hard USD spend caps (MAX_USD_BUDGET), tracking step-by-step token expenditure and denying execution when financial limits are met.
• human-gate (PreToolUse): Event-driven Human-in-the-Loop gating intercepting file mutations to .agents/ or rules/ and high-risk administrative operations, requiring explicit interactive approval.
• thinking-budget (PreToolUse): Intercepts tool calls and reasoning turns to dynamically budget and cap Gemini 3.8 extended reasoning token runaways.
• drift-detector (PreToolUse): Intercepts tool call arguments to detect adversarial prompt injection patterns, suspicious command evasion, and unexpected schema drift.
• context-compactor (PreToolUse): Evaluates tool input payload sizes and trajectory context pressure to recommend AST pruning or compacting oversized buffers.
• zero-trust-gate (PreToolUse): Deterministic container confinement sentinel blocking privileged flags (--privileged), dangerous capabilities (CAP_SYS_ADMIN), and unsafe docker socket mounts.

3. Rules Hierarchy (rules/)

Antigravity automatically discovers rules by walking up from the current file's directory to the repository root:

• rules/common/GEMINI.md: Foundational instructions for precision, minimal changes, and strict verification.
• rules/common/AGENTS.md: Multi-agent interaction protocols, dogfooding guidelines, and progressive disclosure discipline.
• rules/languages/typescript.md: Strict TypeScript, Zod validations, no any, modern ESM.
• rules/languages/python.md: Python 3.12+, type hints, Pydantic v2, Ruff formatting.
• rules/languages/golang.md: Idiomatic Go guidelines, table-driven unit tests, and structured logging.
• rules/paradigms/security-hardened.md: Defense-in-depth, zero-trust file access, credential hygiene.
• rules/paradigms/tdd.md: Enforces strict Red-Green-Refactor test cycle before writing production code.

---

🖥️ The 4 Antigravity Surfaces

Antigravity operates across four interconnected surfaces. everything-gemini supports all of them seamlessly:

+-----------------------------------------------------------------------------------+
|                                 GOOGLE ANTIGRAVITY                                |
+-------------------------+-------------------------+-------------------------------+
|  1. Antigravity 2.0     |  2. Antigravity IDE     |  3. Antigravity CLI ('agy')   |
|  (Desktop App)          |  (VS Code Fork)         |  (Terminal & Headless)        |
|  - Unified Left Sidebar |  - Antigravity Tab      |  - Fast interactive TUI       |
|  - Chat Canvas          |    (Autocomplete/Super) |  - Headless scripting         |
|  - Auxiliary Panes      |  - Inline (Cmd+I)       |  - Automated CI pipes         |
|    (Subagents, Artifacts|  - Code Lenses & Diffs  |  - Settings JSON              |
+-------------------------+-------------------------+-------------------------------+
|                      4. Python SDK ('google-antigravity')                         |
|  - Programmatic leasing  - Streaming thoughts deltas  - Custom tool wiring        |
+-----------------------------------------------------------------------------------+

1. Antigravity 2.0 (Desktop Application): The flagship Electron desktop application. Supports scheduled tasks (cron/timers), background process monitoring, visual artifact viewers, and slash commands.
2. Antigravity IDE: Standalone AI-first integrated editor based on VS Code. Features Antigravity Tab (next-intent prediction), Inline Edits (Cmd+I), Code Lenses, and Diagnostic Auto-Fix.
3. Antigravity CLI (agy): High-speed terminal TUI for keyboard-driven workflows and headless script automation.
4. Antigravity Python SDK (google-antigravity): Public Python SDK allowing developers to build custom agents, stream reasoning tokens (response.thoughts), and wire custom tools.

---

🪄 Slash Commands Playbook

Antigravity provides built-in slash commands in chat that activate specialized cognitive modes:

Command	Purpose	How Everything Gemini Enhances It
/goal <task>	Long-running autonomous mode	Pairs with Everything Gemini's Goal Verifier Hook to prevent premature termination until test suites pass.
/plan <task>	Step-by-step implementation blueprint	Enforces architectural review, risk assessment, and mermaid diagrams before code edits.
/grill-me	Interactive design interview	Asks hard architectural and trade-off questions to resolve ambiguities before implementing.
/learn	Persist feedback across sessions	Automatically indexes insights into a permanent Knowledge Item (<appDataDir>/knowledge).
/schedule	One-shot or recurring background tasks	Sets up cron monitors or health-check agents with reactive wakeup.

---

📊 Token Efficiency & Benchmark

Traditional agent harnesses dump dozens of prompt files into the system prompt on every single turn. Antigravity's Progressive Disclosure paired with Everything Gemini avoids this:

[Traditional Setup]
System Prompt: Base Instructions + All 30 Skills (Full Text) + All Rules = ~48,000 Tokens / turn

[Everything Gemini on Antigravity]
System Prompt: Base Instructions + 30 Skill Summaries (YAML) + Directory Rules = ~3,400 Tokens / turn
Skill Body: Loaded on-demand ONLY when invoked = ~1,500 Tokens (one-time)
Savings: > 85% Token Reduction & 3x Faster First-Token Latency!

Run the built-in diagnostic benchmark across all 158 ecosystem components (117 progressive skills, 28 scoped subagents, 13 lifecycle hooks):

./bin/everything-gemini test-all

---

🛠️ Usage & CLI Reference

The zero-dependency everything-gemini CLI manages your configurations effortlessly:

# Verify your installation and check for conflicts
./bin/everything-gemini doctor

# Inspect active workspace status dashboard (installed skills, subagents, hooks)
./bin/everything-gemini status
./bin/everything-gemini status --global
./bin/everything-gemini status --json

# Compare installed workspace components against master catalog
./bin/everything-gemini diff

# Export installed configuration bundle into a portable JSON profile
./bin/everything-gemini export
./bin/everything-gemini export --output my-profile.json

# Import portable configuration bundle into workspace
./bin/everything-gemini import my-profile.json
./bin/everything-gemini import my-profile.json --dry-run
./bin/everything-gemini import my-profile.json --workspace ./my-project

# Package offline distribution bundle for air-gapped environments or CI/CD
./bin/everything-gemini bundle
./bin/everything-gemini bundle --output ./offline-bundle
./bin/everything-gemini bundle --dry-run
./bin/everything-gemini bundle --json

# Update installed ecosystem components in-place to latest catalog versions
./bin/everything-gemini update
./bin/everything-gemini update --global
./bin/everything-gemini update --dry-run

# Preview an installation without modifying disk
./bin/everything-gemini install --profile developer --dry-run

# Install a specific profile into the current workspace (.agents/)
./bin/everything-gemini install --profile developer

# Install globally for all projects (~/.gemini/config/)
./bin/everything-gemini install --global --profile core

# List all available skills, rules, hooks, and plugins
./bin/everything-gemini list

# Initialize everything-gemini in current workspace
./bin/everything-gemini init

# Add a specific skill to your active workspace
./bin/everything-gemini add skill tdd-workflow

# Remove a skill from your active workspace
./bin/everything-gemini remove skill tdd-workflow

# Add or remove lifecycle safety hooks
./bin/everything-gemini add hook safety-shield
./bin/everything-gemini remove hook safety-shield

# Add a specific plugin
./bin/everything-gemini add plugin web-fullstack

# Search components across skills, subagents, hooks, and MCP blueprints
./bin/everything-gemini search kafka
./bin/everything-gemini search dbt --json

# Run comprehensive ecosystem audit across frontmatter, contracts, hooks, and links
./bin/everything-gemini audit
./bin/everything-gemini audit --json

# Benchmark and validate all 84 skills, 16 subagents, and 12 hooks (112 components)
./bin/everything-gemini test-all
./bin/everything-gemini benchmark --json

# Pre-flight predictive token cost modeling & FinOps spend budgeting across models
./bin/everything-gemini estimate
./bin/everything-gemini estimate --model flash --turns 5 --max-usd 5.00
./bin/everything-gemini estimate --json

# Run Harness³ Autonomous Darwinian genetic self-evolution & benchmark
./bin/everything-gemini harness3 auto-bench --generations 3 --max-usd 2.50
./bin/everything-gemini harness3 --dry-run
./bin/everything-gemini harness3 --json

# Launch interactive visual mission control web dashboard
./bin/everything-gemini ui --port 4200
./bin/everything-gemini ui --dry-run
./bin/everything-gemini ui --json

# Spawn dynamic ephemeral micro-agent swarms with Byzantine consensus
./bin/everything-gemini swarm spawn "refactor auth middleware to zero-trust mTLS" --workers 3
./bin/everything-gemini swarm --dry-run
./bin/everything-gemini swarm --json

# Profile cognitive task complexity and allocate Gemini 3.8 thinking tokens
./bin/everything-gemini thinking profile "Resolve distributed Spanner deadlock"
./bin/everything-gemini thinking budget --level deep --dry-run
./bin/everything-gemini thinking --json

# Run Harness³ autonomous mutation testing, fault injection & resilience scoring
./bin/everything-gemini resilience run --iterations 10
./bin/everything-gemini resilience inject --faults latency,schema,oom --dry-run
./bin/everything-gemini resilience --json

# Inspect OpenTelemetry GenAI semantic spans & distributed traces
./bin/everything-gemini telemetry status
./bin/everything-gemini telemetry --trace-id 4bf92f3577b34da6a3ce929d0e0e4736 --dry-run
./bin/everything-gemini telemetry --json

# Synchronize distributed cross-repository multi-agent swarms with 2PC consensus
./bin/everything-gemini cross-repo status
./bin/everything-gemini cross-repo sync --repos "backend,frontend,infra" --dry-run
./bin/everything-gemini cross-repo --json

# Perform AST-guided context distillation and semantic prompt compaction
./bin/everything-gemini compact analyze
./bin/everything-gemini compact distill --threshold 75 --ast --dry-run
./bin/everything-gemini compact --json

# Synthesize 8-surface Google Research Harness² workspace (prompts, guardrails, memory, etc.)
./bin/everything-gemini harness2
./bin/everything-gemini harness2 ./custom-harness --dry-run
./bin/everything-gemini export --format harness2 --output ./harness-export

# Run recursive Harness² optimization over local tasks with hard spend limits & HITL
./bin/everything-gemini harness2 run --dir ./tasks --bench local --max-usd 15.00 --hitl --snapshot
./bin/everything-gemini harness2 run --dry-run
./bin/everything-gemini harness2 run --json

# Ingest production telemetry failure trace into automated regression probe
./bin/everything-gemini harness2 ingest-trace failure-trace.json --dir ./tasks/regression

# Synthesize Google AX (Agent Executor) cluster manifests (Task, Workspace, Gateway, Model)
./bin/everything-gemini ax init --output ./ax-deploy --name cloud-agent --model gemini-3.8-flash
./bin/everything-gemini ax init --dry-run
./bin/everything-gemini ax init --json

# Federate and sync official Google skills (google/skills) & plugins
./bin/everything-gemini skills sync google/skills
./bin/everything-gemini skills sync --dry-run
./bin/everything-gemini skills sync --json

# Autonomous Company OS & C-Suite Boardroom Intelligence (inspired by agency-agents)
./bin/everything-gemini agency status
./bin/everything-gemini agency debate "Zero-Trust Enterprise Cloud Platform"
./bin/everything-gemini agency debate "Multi-tenant SaaS" --dry-run
./bin/everything-gemini agency veto-check
./bin/everything-gemini agency status --json

# Visualize multi-agent interaction and dependency graph (terminal, Mermaid, or JSON)
./bin/everything-gemini graph
./bin/everything-gemini graph --mermaid
./bin/everything-gemini graph --json

# View detailed playbook or specification for any skill, subagent, or hook
./bin/everything-gemini info agent-evals
./bin/everything-gemini info platform-engineer
./bin/everything-gemini info gemini-shield

# Synchronize workspace with latest catalog & verify integrity
./bin/everything-gemini sync
./bin/everything-gemini sync --dry-run
./bin/everything-gemini sync --json

# Scaffold a new progressive-disclosure skill, scoped subagent, or lifecycle hook
./bin/everything-gemini scaffold skill my-custom-skill
./bin/everything-gemini scaffold agent my-worker-agent
./bin/everything-gemini scaffold hook my-safety-hook

# Generate shell autocompletion script (bash/zsh)
./bin/everything-gemini completion >> ~/.zshrc

# Or automatically install autocompletion idempotently into your active shell rc
./bin/everything-gemini completion --install
./bin/everything-gemini completion --install --dry-run
./bin/everything-gemini completion --install --rc ~/.bashrc

# Validate all schemas, hooks, and frontmatter
./bin/everything-gemini validate

---

🤝 Contributing

We welcome community contributions! You can contribute new Skills, Rules, Lifecycle Hooks, MCP Blueprints, or Plugins.

Check out our Community Wishlist & Good First Issues for high-impact skills ready to be built. See CONTRIBUTING.md for full development guidelines, YAML frontmatter standards, and validation procedures.

---

🌟 Star History

Star History Chart

---

🙏 Acknowledgements & Inspiration

• Inspired by Affan Mustafa's Everything Claude Code (ECC), re-engineered from the ground up for Google Antigravity (AGY) and the Gemini frontier ecosystem.
• Built with 💙 for the Google Antigravity & Gemini developer community.

---

📄 License

This project is licensed under the Apache 2.0 License.