evansenter · evansenter · Feb 7, 2026 · Feb 7, 2026 · Feb 7, 2026 · Feb 7, 2026
diff --git a/.github/workflows/rust.yml b/.github/workflows/rust.yml
@@ -20,8 +20,8 @@ jobs:
       - uses: Swatinem/rust-cache@v2
       - name: Install mold linker
         run: sudo apt-get update && sudo apt-get install -y mold
-      - name: Check workspace
-        run: cargo check --workspace --all-targets --all-features
+      - name: Check
+        run: cargo check --all-targets --all-features
 
   test:
     name: Test Suite
@@ -34,8 +34,7 @@ jobs:
       - name: Install mold linker
         run: sudo apt-get update && sudo apt-get install -y mold
       - name: Run unit tests
-        run: cargo nextest run --workspace
-      # Note: No doctests - clemini is a binary crate without a library target
+        run: cargo nextest run
 
   test-integration:
     name: Integration Tests
@@ -86,7 +85,7 @@ jobs:
       - name: Install mold linker
         run: sudo apt-get update && sudo apt-get install -y mold
       - name: Run clippy
-        run: cargo clippy --workspace --all-targets --all-features -- -D warnings
+        run: cargo clippy --all-targets --all-features -- -D warnings
 
   doc:
     name: Documentation
@@ -98,7 +97,7 @@ jobs:
       - name: Install mold linker
         run: sudo apt-get update && sudo apt-get install -y mold
       - name: Check documentation
-        run: cargo doc --workspace --no-deps --document-private-items
+        run: cargo doc --no-deps --document-private-items
         env:
           RUSTDOCFLAGS: -D warnings
 
@@ -111,8 +110,8 @@ jobs:
       - uses: Swatinem/rust-cache@v2
       - name: Install mold linker
         run: sudo apt-get update && sudo apt-get install -y mold
-      - name: Check workspace with MSRV
-        run: cargo check --workspace
+      - name: Check with MSRV
+        run: cargo check
 
   coverage:
     name: Code Coverage
@@ -127,7 +126,7 @@ jobs:
       - name: Install mold linker
         run: sudo apt-get update && sudo apt-get install -y mold
       - name: Generate coverage
-        run: cargo llvm-cov --workspace --lcov --output-path lcov.info
+        run: cargo llvm-cov --lcov --output-path lcov.info
       - uses: codecov/codecov-action@v4
         with:
           files: lcov.info

diff --git a/.gitignore b/.gitignore
@@ -1,6 +1,5 @@
 # Rust
 /target/
-Cargo.lock
 
 # IDE
 .idea/

diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -5,6 +5,13 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [Unreleased]
+
+### Changed
+- Extracted clemitui into standalone repository ([evansenter/clemitui](https://github.com/evansenter/clemitui)), now referenced as a git dependency
+- Committed Cargo.lock for reproducible builds
+- Removed single-member workspace wrapper
+
 ## [0.4.0] - 2026-01-24
 
 ### Added

diff --git a/CLAUDE.md b/CLAUDE.md
@@ -33,19 +33,7 @@ Logs are stored in `~/.clemini/logs/` with daily rotation.
 
 The CLI has three modes: single-prompt (`-p "prompt"`), interactive REPL, and MCP server (`--mcp-server`).
 
-### Workspace Structure
-
-This project is a Cargo workspace with two crates:
-
-```
-.
-├── Cargo.toml       # Workspace root
-├── src/             # clemini crate (AI agent)
-└── crates/
-    └── clemitui/    # TUI library crate (reusable by any ACP agent)
-```
-
-#### clemini (AI Agent)
+### Source Structure
 
 ```
 src/
@@ -69,27 +57,9 @@ src/
     └── ...          # Individual tool modules (edit, read, grep, etc.)
 ```
 
-#### clemitui (TUI Library)
-
-Standalone crate for terminal UI, usable by any ACP-compatible agent:
-
-```
-crates/clemitui/
-├── Cargo.toml
-├── src/
-│   ├── lib.rs       # Re-exports
-│   ├── format.rs    # Primitive formatting functions (tool output, warnings)
-│   ├── logging.rs   # OutputSink trait, log_event functions
-│   └── text_buffer.rs # TextBuffer for streaming markdown
-└── tests/
-    ├── common/mod.rs        # Shared test helpers (strip_ansi, RAII guards, CaptureSink)
-    ├── acp_simulation_tests.rs  # 29 tests simulating ACP agent patterns
-    └── e2e_tests.rs         # 19 PTY-based tests for actual terminal output
-```
-
-**Design**: clemitui takes primitive types (strings, durations, token counts), not genai-rs types. This allows it to work with any ACP agent. clemini's format.rs re-exports these and adds genai-rs-specific wrappers.
+### clemitui (External TUI Library)
 
-Run clemitui tests: `cargo test -p clemitui`
+Terminal UI is provided by [clemitui](https://github.com/evansenter/clemitui), a standalone crate referenced as a git dependency. It takes primitive types (strings, durations, token counts), not genai-rs types, so it can work with any ACP agent. clemini's `format.rs` and `logging.rs` re-export clemitui's API and add genai-rs-specific wrappers.
 
 ### Event-Driven Architecture
 
@@ -158,6 +128,9 @@ Debugging: `LOUD_WIRE=1` logs all HTTP requests/responses.
 - `GEMINI_API_KEY` - Required
 - Model: `gemini-3-flash-preview`
 - Config: `~/.clemini/config.toml` (optional)
+  - `model` - Gemini model to use (default: `gemini-3-flash-preview`)
+  - `bash_timeout` - Timeout in seconds for bash commands (default: 120)
+  - `allowed_paths` - Additional paths tools can access beyond cwd (default: none)
 
 ## Documentation
 
@@ -169,9 +142,10 @@ Debugging: `LOUD_WIRE=1` logs all HTTP requests/responses.
 
 ## Conventions
 
-- Rust 2024 edition (let chains, etc.)
+- Rust 2024 edition (let chains, etc.), MSRV 1.88 (enforced in CI)
 - Tools return JSON: success data or `{"error": "..."}`
 - Tool errors return as JSON (not propagated) so Gemini can see them and retry
+- CI uses `cargo-nextest` for test execution (`make test-all` uses it locally too)
 
 ## Development Process
 
@@ -189,6 +163,7 @@ Debugging: `LOUD_WIRE=1` logs all HTTP requests/responses.
 - `make clippy` (no warnings)
 - `make fmt` (run formatter, then commit any changes it makes)
 - `make test` (tests pass)
+- Documentation must compile without warnings (CI runs `cargo doc --no-deps --document-private-items` with `-D warnings`)
 
 Don't skip tests. If a test is flaky or legitimately broken by your change, fix the test as part of the PR.
 
@@ -216,7 +191,7 @@ Run locally with: `cargo test --test <name> -- --include-ignored --nocapture`
 These use `validate_response_semantically()` from `tests/common/mod.rs` - a second Gemini call with structured output that judges whether responses are appropriate. This provides a middle ground between brittle string assertions and purely structural checks.
 
 **Shared test helpers** - Common patterns for test utilities:
-- Put shared helpers in `tests/common/mod.rs` (for clemini) or `crates/clemitui/tests/common/mod.rs` (for clemitui)
+- Put shared helpers in `tests/common/mod.rs`
 - Use `#![allow(dead_code)]` in shared test modules since not all test files use all helpers
 - RAII guards for cleanup: `DisableColors` (reset color override on drop), `LoggingGuard` (disable logging on drop)
 - Pattern: `let _guard = DisableColors::new();` at test start ensures cleanup even on panic