chati-dev 2.1.2 → 3.0.1

package/README.md

@@ -10,6 +10,7 @@
   <a href="https://www.npmjs.com/package/chati-dev"><img src="https://img.shields.io/npm/v/chati-dev?color=blue&label=npm" alt="npm"></a>
   <a href="LICENSE"><img src="https://img.shields.io/badge/License-MIT-blue.svg" alt="License: MIT"></a>
   <a href="https://nodejs.org/"><img src="https://img.shields.io/badge/node-%3E%3D20.0.0-brightgreen.svg" alt="Node.js"></a>
+  <a href="#multi-cli-support"><img src="https://img.shields.io/badge/LLM-Claude%20%7C%20Gemini%20%7C%20Codex%20%7C%20Copilot-blueviolet.svg" alt="Multi-LLM"></a>
   <a href="#internationalization"><img src="https://img.shields.io/badge/i18n-EN%20%7C%20PT%20%7C%20ES%20%7C%20FR-informational.svg" alt="i18n"></a>
   <a href=".github/CONTRIBUTING.md"><img src="https://img.shields.io/badge/contributions-welcome-brightgreen.svg" alt="Contributions Welcome"></a>
 </p>
@@ -39,11 +40,11 @@ AI-assisted development today has three critical issues:
 
 1. **Context Loss** — AI forgets decisions across sessions, leading to inconsistent implementations
 2. **Planning Gaps** — Jumping straight to code without structured requirements leads to rework
-3. **System Leakage** — Users accidentally "fall out" of agent systems into generic AI mode
+3. **Vendor Lock-in** — Most AI dev tools only work with one provider, limiting flexibility
 
 ### The Solution
 
-Chati.dev introduces **Agent-Driven Development**: a pipeline where each agent owns a specific phase, produces validated artifacts, and hands off context to the next agent. The system ensures context is never lost, knowledge persists across sessions, and you never accidentally leave the pipeline.
+Chati.dev introduces **Agent-Driven Development**: a pipeline where each agent owns a specific phase, produces validated artifacts, and hands off context to the next agent. The system ensures context is never lost, knowledge persists across sessions, and you can use any supported AI provider.
 
 ---
 
@@ -52,9 +53,11 @@ Chati.dev introduces **Agent-Driven Development**: a pipeline where each agent o
 ### Prerequisites
 
 - **Node.js** >= 20.0.0 ([download](https://nodejs.org/))
-- **Claude Code** — the CLI for Claude ([install guide](https://docs.anthropic.com/en/docs/claude-code/overview))
-
-Other supported IDEs: VS Code, Cursor, Gemini CLI, GitHub Copilot, AntiGravity (see [Supported IDEs](#supported-ides)).
+- **An AI CLI** — at least one of:
+  - [Claude Code](https://docs.anthropic.com/en/docs/claude-code/overview) (recommended)
+  - [Gemini CLI](https://github.com/google-gemini/gemini-cli)
+  - [Codex CLI](https://github.com/openai/codex)
+  - [GitHub Copilot CLI](https://docs.github.com/en/copilot)
 
 ### 1. Install in your project
 
@@ -64,11 +67,18 @@ Open a terminal in your project directory:
 npx chati-dev init
 ```
 
-The wizard asks your language and project type, then auto-configures everything.
+The wizard guides you through 4 steps:
+
+| Step | What it asks |
+|------|-------------|
+| **Language** | English, Portuguese, Spanish, or French |
+| **Project Type** | Greenfield (new) or Brownfield (existing) |
+| **AI Provider** | Claude, Gemini, Codex, or Copilot — auto-configures optimal models per agent |
+| **Confirm** | Review summary and proceed |
 
 ### 2. Activate the orchestrator
 
-Open Claude Code in the same project directory, then type:
+Open your AI CLI in the same project directory, then type:
 
 ```
 /chati
@@ -110,14 +120,74 @@ npx chati-dev status --watch  # Auto-refresh every 5s
 | Feature | What it means |
 |---------|--------------|
 | **13 Specialized Agents** | Each agent has a defined mission, success criteria, and handoff protocol — not one AI trying to do everything |
-| **Quality Gates** | Every phase is validated before moving forward. Binary pass/fail — no subjective "looks good" |
+| **Multi-CLI Architecture** | Choose your AI provider at install time: Claude, Gemini, Codex, or Copilot. Each agent gets the optimal model for that provider |
+| **Quality Gates** | Every phase is validated before moving forward. 3-tier verdicts: APPROVED, NEEDS_REVISION, or BLOCKED |
 | **Context Persistence** | Sessions survive restarts. Close your IDE, come back next week — the system remembers everything |
 | **Session Lock** | Once activated, you stay inside the system. No accidentally "falling out" into generic AI mode |
 | **Multi-Terminal** | Autonomous agents run in parallel in separate terminals. Detail, Architect, and UX agents work simultaneously |
 | **Memory System** | The system learns from mistakes. Gotchas are captured automatically and recalled when relevant |
+| **Execution Profiles** | Three profiles — explore (read-only), guided (default), autonomous (gate >= 90%) — with safety net and circuit breaker |
 | **IDE-Agnostic** | Works with Claude Code, VS Code, Cursor, Gemini CLI, GitHub Copilot, and AntiGravity |
 | **4 Languages** | Interface supports English, Portuguese, Spanish, and French. Artifacts are always generated in English |
-| **Supply Chain Security** | Every file is cryptographically signed. Tampered packages are blocked on install |
+| **Supply Chain Security** | Every file is cryptographically signed (Ed25519). Tampered packages are blocked on install |
+
+---
+
+## Multi-CLI Support
+
+Chati.dev is provider-agnostic. You choose your AI provider during installation, and the system auto-configures optimal model assignments for every agent.
+
+### Supported Providers
+
+| Provider | CLI | Deep Reasoning | Lightweight | Status |
+|----------|-----|---------------|-------------|--------|
+| **Claude** | `claude` | Opus | Sonnet / Haiku | Default |
+| **Gemini** | `gemini` | Pro | Flash | Full support |
+| **Codex** | `codex` | Codex | Codex Mini | Full support |
+| **Copilot** | `copilot` | Sonnet 4.5 | GPT-5 | Full support |
+
+### How Model Assignment Works
+
+Each agent is classified by its reasoning needs:
+
+- **Deep reasoning** (Architect, QA, Dev, Detail, Brownfield-WU) → top-tier model
+- **Lightweight** (Brief, Phases, UX, Greenfield-WU, DevOps, Orchestrator) → fast model
+
+When you select a provider, the system maps each agent to the best available model:
+
+```
+Claude:  architect → opus,   brief → sonnet,  greenfield-wu → haiku
+Gemini:  architect → pro,    brief → flash,   greenfield-wu → flash
+Codex:   architect → codex,  brief → codex-mini
+Copilot: architect → claude-sonnet, brief → claude-sonnet
+```
+
+### CLI Invocation Syntax
+
+Each provider uses its own CLI syntax. Chati.dev handles this transparently:
+
+| Provider | Command | Example |
+|----------|---------|---------|
+| Claude | `claude --print --model <id>` | `claude --print --model claude-opus-4-6` |
+| Gemini | `gemini --model <id>` | `gemini --model gemini-2.5-pro` |
+| Codex | `codex exec -m <id>` | `codex exec -m gpt-5.3-codex` |
+| Copilot | `copilot -p --model <id>` | `copilot -p --model claude-sonnet-4.5` |
+
+Prompts are piped via stdin for all providers. You can override individual agent models in `chati.dev/config.yaml` under `agent_overrides`.
+
+---
+
+## Execution Profiles
+
+Three profiles control how much autonomy agents have:
+
+| Profile | Behavior | When to use |
+|---------|----------|-------------|
+| **explore** | Read-only. Agents analyze but don't modify files | Understanding a new codebase |
+| **guided** | Default. Agents propose changes, you approve | Normal development workflow |
+| **autonomous** | Agents execute without confirmation (quality gates >= 90%) | Trusted pipelines with high quality scores |
+
+The system starts in `guided` mode. Transition to `autonomous` requires both QA gates scoring >= 95%. A safety net with 5 triggers (stuck loop, quality drop, scope creep, error cascade, user override) automatically reverts to guided mode when needed.
 
 ---
 
@@ -131,7 +201,7 @@ npx chati-dev status --watch  # Auto-refresh every 5s
 | **PLAN** | Detail, Architect, UX, Phases, Tasks | Create PRD, design architecture, define UX, break work into phases and tasks |
 | **BUILD** | Dev | Implement code task by task, following the plan |
 | **DEPLOY** | DevOps | Handle git operations, deployment, and documentation |
-| **Quality** | QA-Planning, QA-Implementation | Validate plan coherence (≥95%) and code quality (≥95%) between phases |
+| **Quality** | QA-Planning, QA-Implementation | Validate plan coherence (>= 95%) and code quality (>= 95%) between phases |
 
 ### How the Pipeline Works
 
@@ -152,6 +222,7 @@ npx chati-dev status --watch  # Auto-refresh every 5s
        │    requirements into a structured doc  │
        │                                        │
        ├─ Spawns separate terminals ────────────┐
+       │   (using your selected AI provider)    │
        │                                        │
        │    PLAN (autonomous, parallel)         │
        │    Detail ───┐                         │
@@ -162,7 +233,7 @@ npx chati-dev status --watch  # Auto-refresh every 5s
        ├─ Quality Gate ─────────────────────────┐
        │                                        │
        │    QA-Planning validates the plan      │
-       │    Must score ≥ 95% to proceed         │
+       │    Must score >= 95% to proceed        │
        │                                        │
        ├─ Spawns terminal(s) ───────────────────┐
        │                                        │
@@ -183,7 +254,7 @@ npx chati-dev status --watch  # Auto-refresh every 5s
                                          ✓ Done │
 ```
 
-Each spawned terminal runs as a separate `claude -p --model <model>` process with its own context window, write-scope isolation, and structured handoff output.
+Each spawned terminal runs as a separate CLI process with its own context window, write-scope isolation, and structured handoff output. The AI provider and model are selected automatically based on your configuration.
 
 ### Intelligence Layer
 
@@ -197,15 +268,16 @@ Three systems operate transparently behind the pipeline:
 
 ### Constitution
 
-The system is governed by a **17-article Constitution** that enforces agent behavior, quality standards, security, and system integrity:
+The system is governed by a **19-article Constitution** that enforces agent behavior, quality standards, security, and system integrity:
 
 - **Agent Governance** — Every agent has a defined mission, scope, and success criteria
-- **Quality Standards** — Minimum 95% score on quality gates. Binary pass/fail, not subjective
-- **Security** — No secrets in system files. No destructive operations without confirmation
+- **Quality Standards** — Minimum 95% score on quality gates. 3-tier verdicts (APPROVED / NEEDS_REVISION / BLOCKED)
+- **Security** — No secrets in system files. No destructive operations without confirmation. Ed25519 supply chain verification
 - **Mode Governance** — Planning mode can't modify project code. Build mode has full access
 - **Session Lock** — Once activated, all messages route through the orchestrator
-- **Model Governance** — Each agent runs on its designated AI model (Opus, Sonnet, or Haiku)
-- **Execution Modes** — Autonomous and human-in-the-loop with safety net and circuit breaker
+- **Model Governance** — Each agent runs on its designated model, enforced by the CLI adapter
+- **Execution Profiles** — Explore, guided, and autonomous modes with safety net and circuit breaker
+- **Multi-CLI** — Provider-agnostic architecture with adapter pattern and automatic model mapping
 
 ---
 
@@ -216,8 +288,8 @@ The system is governed by a **17-article Constitution** that enforces agent beha
 | **Claude Code** | `.claude/commands/chati.md` → orchestrator |
 | **VS Code** | `.vscode/chati.md` → orchestrator |
 | **Cursor** | `.cursor/rules/chati.md` → orchestrator |
-| **Gemini CLI** | `.gemini/agents/chati.md` → orchestrator |
-| **GitHub Copilot** | `.github/copilot/chati.md` → orchestrator |
+| **Gemini CLI** | `.gemini/commands/chati.toml` → orchestrator |
+| **GitHub Copilot** | `.github/agents/chati.md` → orchestrator |
 | **AntiGravity** | Platform agent config → orchestrator |
 
 All IDEs use a thin router file that points to the same orchestrator. Your project works the same regardless of which IDE you use.
@@ -291,7 +363,6 @@ your-project/
 │   │   ├── quality/              # QA-Planning, QA-Implementation
 │   │   ├── build/                # Dev
 │   │   └── deploy/               # DevOps
-│   ├── tasks/                    # 73 task definitions
 │   ├── workflows/                # 6 workflow blueprints
 │   ├── templates/                # 6 artifact templates
 │   ├── schemas/                  # 5 JSON schemas
@@ -305,7 +376,7 @@ your-project/
 │   ├── data/                     # Entity registry
 │   ├── i18n/                     # EN, PT, ES, FR translations
 │   ├── migrations/               # Version migration scripts
-│   ├── constitution.md           # 17 Articles + Preamble
+│   ├── constitution.md           # 19 Articles + Preamble
 │   └── config.yaml               # System configuration
 └── packages/
     └── chati-dev/                # CLI + runtime engine
@@ -320,9 +391,9 @@ The installer and agent interactions support 4 languages:
 | Language | Code | Status |
 |----------|------|--------|
 | **English** | `en` | Default |
-| **Português** | `pt` | Full support |
-| **Español** | `es` | Full support |
-| **Français** | `fr` | Full support |
+| **Portugues** | `pt` | Full support |
+| **Espanol** | `es` | Full support |
+| **Francais** | `fr` | Full support |
 
 Artifacts are always generated in English for portability and team collaboration.
 

package/bin/chati.js

@@ -255,7 +255,7 @@ async function main() {
       console.log('  - 13 agents (orchestrator + 12 specialized)');
       console.log('  - 6 workflow blueprints');
       console.log('  - 6 templates');
-      console.log('  - Constitution (17 Articles + Preamble)');
+      console.log('  - Constitution (19 Articles + Preamble)');
       console.log('  - Dashboard TUI');
       console.log('  - Upgrade system with migrations');
       console.log('  - 6 IDE support');

package/framework/agents/build/dev.md

@@ -14,6 +14,7 @@ You are the **Dev Agent**, responsible for implementing code based on the approv
 - **Ratio**: 20% Human / 80% AI (interactive) or 5% Human / 95% AI (autonomous)
 - **Absorbs**: Dev personas, self-validation patterns, Design System token enforcement, Ralph Wiggum autonomous mode, blocker taxonomy
 - **Model**: opus | no downgrade (code generation requires highest quality)
+- **Provider**: claude (default) | gemini (when large codebase tasks)
 
 ## Required MCPs
 - context7 (library documentation)

package/framework/agents/deploy/devops.md

@@ -13,6 +13,7 @@ You are the **DevOps Agent**, responsible for shipping the project: Git operatio
 - **Duration**: 15-30 min
 - **Ratio**: 30% Human / 70% AI
 - **Model**: sonnet | upgrade: opus if multi-environment or infrastructure-as-code
+- **Provider**: claude (default)
 - **Absorbs**: Docs-gen (auto documentation generation)
 
 ## Required MCPs

package/framework/agents/discover/brief.md

@@ -13,6 +13,7 @@ You are the **Brief Agent**, responsible for extracting and structuring the core
 - **Duration**: 30-60 min
 - **Ratio**: 90% Human / 10% AI
 - **Model**: sonnet | upgrade: opus if enterprise or 10+ integrations
+- **Provider**: claude (default)
 
 ## Required MCPs
 - None

package/framework/agents/discover/brownfield-wu.md

@@ -13,6 +13,7 @@ You are the **Brownfield WU Agent**, responsible for comprehensive discovery of
 - **Duration**: 1-4h depending on codebase size
 - **Ratio**: 40% Human / 60% AI
 - **Model**: opus | no downgrade (codebase analysis requires deep reasoning)
+- **Provider**: claude (default) | gemini (when codebase > 100K LOC)
 
 ## Required MCPs
 - git (read-only)

package/framework/agents/discover/greenfield-wu.md

@@ -13,6 +13,7 @@ You are the **Greenfield WU Agent**, responsible for understanding HOW the user
 - **Duration**: 15-30 min
 - **Ratio**: 70% Human / 30% AI
 - **Model**: haiku | upgrade: sonnet if multi-stack or enterprise
+- **Provider**: claude (default)
 
 ## Required MCPs
 - None

package/framework/agents/plan/architect.md

@@ -14,6 +14,7 @@ You are the **Architect Agent**, responsible for defining HOW the system will be
 - **Ratio**: 50% Human / 50% AI
 - **Absorbs**: Data Engineer (data modeling, DB audit, schema design, RLS policies)
 - **Model**: opus | no downgrade (architecture decisions require deep reasoning)
+- **Provider**: claude (default)
 
 ## Required MCPs
 - context7 (library documentation search)

package/framework/agents/plan/detail.md

@@ -14,6 +14,7 @@ You are the **Detail Agent**, responsible for transforming the Brief into a form
 - **Ratio**: 70% Human / 30% AI
 - **Absorbs**: PM (PRD creation, product strategy), Analyst (market research, competitive analysis)
 - **Model**: opus | no downgrade (PRD traceability requires deep reasoning)
+- **Provider**: claude (default)
 
 ## Required MCPs
 - None

package/framework/agents/plan/phases.md

@@ -1,6 +1,6 @@
 # Phases Agent — Roadmap & Wave Planning
 
-You are the **Phases Agent**, responsible for breaking the PRD into development phases with an MVP-first approach. You absorb the SM (Scrum Master) responsibilities for sprint planning, prioritization, and sequencing.
+You are the **Phases Agent**, responsible for breaking the PRD into development phases with an MVP-first approach. You absorb the SM (Scrum Master) responsibilities for phase planning, prioritization, and sequencing.
 
 ---
 
@@ -12,8 +12,9 @@ You are the **Phases Agent**, responsible for breaking the PRD into development
 - **Question Answered**: WHEN will we build it?
 - **Duration**: 20-40 min
 - **Ratio**: 60% Human / 40% AI
-- **Absorbs**: SM (sprint planning, prioritization, wave sequencing)
+- **Absorbs**: SM (phase planning, prioritization, wave sequencing)
 - **Model**: sonnet | upgrade: opus if 20+ requirements or complex dependencies
+- **Provider**: claude (default)
 
 ## Required MCPs
 - None

package/framework/agents/plan/tasks.md

@@ -14,6 +14,7 @@ You are the **Tasks Agent**, responsible for breaking phases into atomic, execut
 - **Ratio**: 50% Human / 50% AI
 - **Absorbs**: SM (structured task creation, Given-When-Then criteria)
 - **Model**: sonnet | upgrade: opus if 50+ tasks or complex acceptance criteria
+- **Provider**: claude (default)
 
 ## Required MCPs
 - None

package/framework/agents/plan/ux.md

@@ -14,6 +14,7 @@ You are the **UX Agent**, responsible for defining HOW the product will look and
 - **Ratio**: 60% Human / 40% AI
 - **Absorbs**: Design System init + audit (embedded workflow)
 - **Model**: sonnet | upgrade: opus if design system creation from scratch
+- **Provider**: claude (default)
 
 ## Required MCPs
 - None

package/framework/agents/quality/qa-implementation.md

@@ -13,6 +13,7 @@ You are the **QA-Implementation Agent**, the quality gate between BUILD and DEPL
 - **Duration**: 15-45 min (mostly automated)
 - **Ratio**: 90% AI / 10% Human
 - **Model**: opus | no downgrade (code review and SAST require deep reasoning)
+- **Provider**: claude (default)
 - **Absorbs**: Tests + SAST + CodeRabbit code review
 
 ## Required MCPs

package/framework/agents/quality/qa-planning.md

@@ -13,6 +13,7 @@ You are the **QA-Planning Agent**, the quality gate between PLANNING (planning)
 - **Duration**: 15-30 min (automated validation)
 - **Ratio**: 95% AI / 5% Human
 - **Model**: opus | no downgrade (cross-artifact validation requires deep reasoning)
+- **Provider**: claude (default)
 - **Absorbs**: Manager (cross-artifact validation), PO (quality gate validation, coherence checks)
 
 ## Required MCPs

package/framework/config.yaml

@@ -1,8 +1,27 @@
 # chati.dev Configuration
-version: "2.1.0"
+version: "3.0.1"
 installed_at: "2026-02-07T10:00:00Z"
 updated_at: "2026-02-17T00:00:00Z"
-installer_version: "2.1.0"
+installer_version: "3.0.1"
 project_type: greenfield
 language: en
 ides: [claude-code]
+
+# CLI providers — multi-CLI is opt-in (Article XIX)
+providers:
+  claude:
+    enabled: true
+    primary: true
+  gemini:
+    enabled: false
+    model_default: pro
+  codex:
+    enabled: false
+    model_default: codex
+  copilot:
+    enabled: false
+
+# Per-agent provider/model overrides (optional — inherits from AGENT_MODELS default)
+# agent_overrides:
+#   dev: { provider: gemini, model: pro }
+#   brownfield-wu: { provider: gemini, model: pro }

package/framework/constitution.md

@@ -17,7 +17,7 @@ Chati.dev is a planning-first AI-assisted orchestration system that coordinates
 ### Anti-Patterns (NEVER do these)
 
 - **Skip Brief**: Jumping to implementation without understanding the problem guarantees solving the wrong problem.
-- **Long Sprints**: Phases longer than 2 weeks lose focus and delay feedback.
+- **Long Phases**: Phases longer than 2 weeks lose focus and delay feedback.
 - **Ignore Feedback**: User corrections during any agent's execution are signals, not interruptions.
 - **Over-engineering**: Building for hypothetical future requirements adds complexity without value.
 - **Excessive Docs**: Documentation should enable action, not demonstrate thoroughness. If nobody reads it, don't write it.
@@ -368,6 +368,18 @@ The orchestrator SHALL select the optimal AI model for each agent to balance qua
 
 8. Model selections are logged in session.yaml under `model_selections[]` for cost tracking and optimization.
 
+9. Four provider categories are recognized for multi-CLI execution:
+   - **claude**: Primary provider. Deep reasoning, complex analysis, code generation. Full hook support, MCP support.
+   - **gemini**: Large-context provider. Codebase analysis, discovery, document review. 1M token context window. Full hook support, MCP support.
+   - **codex**: Rapid-coding provider. Fast code generation, sandbox execution. No hook support, MCP support.
+   - **copilot**: Multi-model provider. Reads CLAUDE.md, GEMINI.md, and AGENTS.md natively. Full hook support, MCP support.
+
+10. Claude is the default and primary provider. Multi-CLI is opt-in via `config.yaml` `providers` section. When only `claude` is enabled, the system behaves identically to v2.x.
+
+11. Provider assignments per agent are defined in `config.yaml` under `agent_overrides`. When no override exists, the agent uses the primary provider.
+
+12. For providers without hook support (e.g., codex), governance is enforced by prompt injection — the PRISM context block is embedded in the prompt rather than injected via hooks. This provides softer but functional governance.
+
 **Enforcement: GUIDE** — Model recommendations are advisory in IDE mode; automatic in SDK mode.
 
 ---
@@ -402,5 +414,58 @@ The system SHALL support two execution modes that govern the degree of human inv
 
 ---
 
-*Chati.dev Constitution v2.1.0 — 17 Articles + Preamble*
+## Article XVIII: Execution Profile Governance
+
+The system SHALL support three execution profiles that govern the degree of confirmation required for agent actions. Profiles are orthogonal to mode governance (Article XI) — modes control WHERE agents can write; profiles control WHETHER confirmation is required. Note: Execution Mode (Article XVII) governs WHO decides (human vs system); Execution Profile governs HOW actions execute (read-only, confirmed, autonomous).
+
+1. Three profiles are recognized:
+   - **explore**: Read-only mode. Agents analyze, investigate, and suggest but perform NO write operations. Useful for safe discovery and analysis.
+   - **guided**: Default profile. Agents propose actions and wait for user confirmation before writes. Balances safety with productivity.
+   - **autonomous**: Full autonomy. Agents execute without confirmation when quality gate scores meet thresholds. Requires cumulative gate scores >= 95%.
+
+2. The default profile is `guided`. Profile selection is stored in `session.yaml` under `execution_profile`.
+
+3. Profile transitions are logged in `session.yaml` under `profile_transitions[]` for audit trail.
+
+4. Certain operations ALWAYS require confirmation regardless of profile:
+   - Destructive operations (file deletion, database drops, force push)
+   - Deployment to production environments
+   - Deviation protocol activation
+   - Backward pipeline transitions (build → planning)
+
+5. The `autonomous` profile is gated by quality: it can only be activated when the most recent QA gate score is >= 95%. If quality drops below threshold during autonomous execution, the system SHALL automatically downgrade to `guided`.
+
+6. The `explore` profile is available at any pipeline position and does not affect mode governance. An agent in `planning` mode with `explore` profile can read any file but write to none.
+
+**Enforcement: BLOCK** — Write operations in `explore` profile are rejected. Autonomous mode without qualifying gate scores is rejected.
+
+---
+
+## Article XIX: Multi-CLI Governance
+
+When multiple CLI providers are enabled, the system SHALL coordinate agent execution across providers while maintaining governance consistency.
+
+1. The CLI Provider Registry (`packages/chati-dev/src/terminal/cli-registry.js`) is the source of truth for provider capabilities: command syntax, model flags, stdin support, hook support, MCP support, and context file format.
+
+2. The handoff format is provider-agnostic. All agents, regardless of which CLI executes them, produce handoffs in the same two-layer format (Article VIII). This ensures seamless inter-provider communication.
+
+3. Provider availability SHALL be validated by the health check engine before spawning. If a configured provider is unavailable, the system SHALL fall back to the primary provider (claude) with a warning.
+
+4. Context file generation is automatic. When a provider is enabled in `config.yaml`, the installer SHALL generate the corresponding context file (GEMINI.md for gemini, AGENTS.md for codex) derived from CLAUDE.md content.
+
+5. Hook-based governance (constitution-guard, mode-governance, read-protection) applies ONLY to providers with hook support. For providers without hooks, equivalent governance is enforced via prompt injection — the PRISM context block includes governance directives.
+
+6. The orchestrator SHALL select the optimal provider for each agent based on:
+   a. Agent's provider preference (defined in agent Identity section)
+   b. Project-level overrides (config.yaml `agent_overrides`)
+   c. Provider availability (health check)
+   d. Fallback: primary provider (claude)
+
+7. Cost tracking SHALL include provider information. Each model selection entry in session.yaml includes: provider, model, agent, timestamp, and estimated token usage.
+
+**Enforcement: GUIDE** — Provider selection is advisory. The system falls back gracefully when providers are unavailable.
+
+---
+
+*Chati.dev Constitution v3.0.0 — 19 Articles + Preamble*
 *All agents are bound by this Constitution. Violations are enforced per article.*

package/framework/context/governance.md

@@ -1,6 +1,6 @@
 # Chati.dev Governance Rules
 
-Extracted from `chati.dev/constitution.md` (17 Articles). Read the full constitution for complete rules.
+Extracted from `chati.dev/constitution.md` (19 Articles). Read the full constitution for complete rules.
 
 ## Mode Governance (Article XI)
 - **3 modes**: planning (read all, write chati.dev/ only), build (full access), deploy (full + infra)
@@ -31,7 +31,20 @@ Extracted from `chati.dev/constitution.md` (17 Articles). Read the full constitu
 - Per-agent model selection: opus (deep reasoning), sonnet (structured), haiku (lightweight)
 - Model recorded in session for cost tracking
 
-## Execution Mode (Article XVII)
-- Autonomous mode requires gate score >= 90% (qa-planning >= 95%, qa-implementation >= 95%)
+## Execution Mode (Article XVII) — WHO decides
+- Controls whether human or system makes pipeline decisions
+- Autonomous mode requires gate score >= 95% (qa-planning >= 95%, qa-implementation >= 95%)
 - Safety net with 5 triggers: stuck loop, quality drop, scope creep, error cascade, user override
 - Circuit breaker: CLOSED -> OPEN (3 failures) -> HALF_OPEN (probe)
+
+## Execution Profile Governance (Article XVIII) — HOW actions execute
+- 3 profiles: explore (read-only), guided (default), autonomous (gate >= 95%)
+- Transition to autonomous requires QA-Planning >= 95% AND QA-Implementation >= 95%
+- Safety net: 5 triggers revert to guided (stuck loop, quality drop, scope creep, error cascade, user override)
+- Circuit breaker: CLOSED -> OPEN (3 failures) -> HALF_OPEN (probe)
+
+## Multi-CLI Governance (Article XIX)
+- 4 providers: Claude, Gemini, Codex, Copilot
+- Provider selected at install time, auto-configures optimal models per agent
+- Resolution chain: agent_overrides > agent default > primary provider
+- Context files auto-generated for non-Claude providers (GEMINI.md, AGENTS.md)

package/framework/context/root.md

@@ -1,10 +1,10 @@
 # Chati.dev System Context
 
 ## Framework
-- **Version**: 2.1.0
+- **Version**: 3.0.0
 - **Agents**: 13 (12 specialized + orchestrator)
-- **Constitution**: 17 Articles + Preamble
-- **Quality**: 5 pipeline gates + 4-tier validation
+- **Constitution**: 19 Articles + Preamble
+- **Quality**: 5 pipeline gates + 3-tier verdicts (APPROVED / NEEDS_REVISION / BLOCKED)
 
 ## Key References
 - **Session State**: `.chati/session.yaml` (runtime — not committed)

package/framework/data/entity-registry.yaml

@@ -257,7 +257,7 @@ entities:
     constitution:
       path: chati.dev/constitution.md
       type: governance
-      purpose: "17 Articles + Preamble governing all agent behavior"
+      purpose: "19 Articles + Preamble governing all agent behavior"
       keywords: [constitution, governance, rules, articles, enforcement]
       dependencies: []
       adaptability: 0.1

package/framework/domains/agents/brownfield-wu.yaml

@@ -47,3 +47,7 @@ rules:
   - id: bf-05
     text: "MUST assess technical debt and provide risk scores in wu-report.yaml"
     priority: normal
+
+  # Provider Preference (v3.0.0)
+  - wu-provider: "Default provider: claude. Gemini recommended for large codebases (>100K LOC) due to 1M context window."
+    priority: normal

package/framework/domains/agents/detail.yaml

@@ -10,7 +10,7 @@ authority:
     - Requirements prioritization (MoSCoW)
   allowed:
     - Expand brief requirements into detailed specifications
-    - Define user stories with acceptance criteria
+    - Define requirements with acceptance criteria
     - Specify performance, security, and scalability requirements
     - Document edge cases and error scenarios
     - Create requirement traceability matrix

package/framework/domains/agents/dev.yaml

@@ -47,3 +47,7 @@ rules:
   - id: dev-iterate
     text: "Iterate on implementation until acceptance criteria are met."
     priority: normal
+
+  # Provider Preference (v3.0.0)
+  - dev-provider: "Default provider: claude. Can run on gemini for large codebase tasks (1M context window)."
+    priority: normal

package/framework/domains/agents/orchestrator.yaml

@@ -47,3 +47,11 @@ rules:
   - id: orch-session
     text: "Persist all state changes to session.yaml immediately."
     priority: high
+
+  # Provider Routing (v3.0.0)
+  - provider-routing: "Select optimal provider for each agent based on: agent preference, project overrides, provider availability, fallback to primary."
+    priority: high
+  - provider-health: "Validate provider availability via health check before spawning agent terminals."
+    priority: high
+  - provider-command: "/chati providers lists all enabled providers and their status."
+    priority: normal

package/framework/domains/agents/phases.yaml

@@ -17,7 +17,7 @@ authority:
   blocked:
     - Code implementation
     - Task-level decomposition (that's tasks agent's role)
-    - Detailed estimation (story points)
+    - Detailed estimation (hours or sizing)
     - Deployment planning
     - Architecture redesign
   redirectMessage: "Phase breakdown is complete. Redirecting to tasks agent for task decomposition."
@@ -27,7 +27,7 @@ outputs:
 
 rules:
   - id: ph-01
-    text: "MUST define Phase 0 as MVP with minimum viable feature set for user validation"
+    text: "MUST define Phase 1 as MVP with minimum viable feature set for user validation"
     priority: critical
 
   - id: ph-02

package/framework/domains/agents/tasks.yaml

@@ -6,7 +6,7 @@ authority:
     - Task decomposition from phases
     - Atomic task definition
     - Task acceptance criteria writing (Given-When-Then)
-    - Task estimation (story points or hours)
+    - Task estimation (hours or t-shirt sizing)
     - Task dependency mapping
   allowed:
     - Break phases into executable tasks
@@ -40,7 +40,7 @@ rules:
     priority: high
 
   - id: tk-04
-    text: "MUST estimate task complexity using consistent scale (story points or hours)"
+    text: "MUST estimate task complexity using consistent scale (hours or t-shirt sizing)"
     priority: normal
 
   - id: tk-05