docs: enrich CLAUDE.md with project conventions

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: ashsolei

CLAUDE.md

@@ -1,7 +1,89 @@
-@AGENTS.md
-## AgentHub
-- Central hub: `~/AgentHub/`
-- Skills: `.agents/skills/` (symlinked to AgentHub shared skills)
-- MCP: 12 servers synced across all agents
-- Agents: 14 shared agents available
-- Hooks: Safety, notification, and logging hooks
+# CLAUDE.md
+
+## Project: OpenShell
+
+**Organization:** AiFeatures (iAiFy Enterprise)
+**Language:** Rust + Python
+**Description:** Safe, private sandboxed runtime for autonomous AI agents with declarative YAML policy enforcement
+
+## Build & Test
+
+```bash
+# Install dependencies (Python)
+uv sync --all-extras --group dev
+
+# Install CLI binary
+uv pip install .
+
+# Build Rust crates
+cargo build
+
+# Test (Python)
+uv run pytest -v --tb=short
+
+# Test (Rust)
+cargo test
+
+# Pre-commit checks (lint + format + license headers)
+mise run pre-commit
+
+# Full local CI
+mise run ci
+
+# Lint (Python)
+uv run ruff check
+
+# Format (Python)
+uv run ruff format
+
+# E2E tests (requires running cluster)
+mise run e2e
+```
+
+## Architecture
+
+```text
+crates/
+  openshell-cli/          # User-facing CLI binary
+  openshell-server/       # Gateway control-plane API
+  openshell-sandbox/      # Container supervision, policy-enforced egress
+  openshell-policy/       # Filesystem, network, process, inference constraints
+  openshell-router/       # Privacy-aware LLM routing
+  openshell-bootstrap/    # K3s cluster setup, mTLS PKI
+  openshell-ocsf/         # OCSF v1.7.0 structured logging
+  openshell-core/         # Common types, config, error handling
+  openshell-providers/    # Credential provider backends
+  openshell-tui/          # Ratatui terminal dashboard
+python/openshell/         # Python SDK and CLI packaging
+proto/                    # Protobuf / gRPC service contracts
+deploy/                   # Docker, Helm, K8s manifests
+.agents/skills/           # Agent workflow automation
+.agents/agents/           # Sub-agent definitions
+architecture/             # Design decisions and component docs
+```
+
+## Conventions
+
+- Conventional commits: `feat:`, `fix:`, `chore:`, `docs:`, `refactor:`, `test:`, `ci:`, `perf:`
+- Kebab-case file names
+- Branch protection on main -- PRs required
+- Always use `mise` commands over direct docker/cargo builds when available
+- Always use `uv` for Python commands
+- Run `mise run pre-commit` before every commit
+- OCSF structured logging for observable sandbox events; plain `tracing` for internal plumbing
+- Never mention AI agents in commit messages (no Co-Authored-By)
+
+## Shared Resources
+
+| Asset | Location |
+| --- | --- |
+| CI/CD workflows | Ai-road-4-You/enterprise-ci-cd@v1 |
+| Composite actions | Ai-road-4-You/github-actions@v1 |
+| Governance | Ai-road-4-You/governance |
+
+## Fork Info
+
+- Upstream: NVIDIA/OpenShell
+- Do NOT create PRs to upstream
+- Sync managed by Ai-road-4-You/fork-sync
+- Divergence is intentional where documented in README iAiFy Fork Notes