Add CLI commands for store, recall, and forget

CLAUDE.md

@@ -15,9 +15,15 @@ cargo run --release
 cargo run --release -- --verbose
 
 # Run CLI commands
-cargo run --release -- init          # Initialize project
-cargo run --release -- stats         # Show database statistics
-cargo run --release -- list          # List stored items
+cargo run --release -- init                     # Initialize project
+cargo run --release -- stats                    # Show database statistics
+cargo run --release -- list                     # List stored items
+cargo run --release -- list --scope global      # List global items
+cargo run --release -- store "some content"     # Store content
+cargo run --release -- store -                  # Store from stdin
+cargo run --release -- recall "search query"    # Semantic search
+cargo run --release -- forget <id>              # Delete an item
+cargo run --release -- --json list              # JSON output (any command)
 
 # Run tests (requires model download on first run)
 cargo test
@@ -38,7 +44,7 @@ Sediment is a semantic memory system for AI agents, running as an MCP (Model Con
 
 ### Core Components
 
-- **`src/main.rs`** - CLI entry point with subcommands (init, stats, list) and MCP server startup
+- **`src/main.rs`** - CLI entry point with subcommands (init, stats, list, store, recall, forget) and MCP server startup
 - **`src/lib.rs`** - Library root exposing public API, project detection, scope types, and project ID migration
 - **`src/db.rs`** - LanceDB wrapper handling vector storage, hybrid search (vector + FTS/BM25), and CRUD operations
 - **`src/embedder.rs`** - Local embeddings using `all-MiniLM-L6-v2` via Candle (384-dim vectors)

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "sediment-mcp"
-version = "0.4.4"
+version = "0.5.0"
 edition = "2024"
 repository = "https://github.com/rendro/sediment"
 homepage = "https://github.com/rendro/sediment"

README.md

@@ -141,12 +141,19 @@ Go to **Settings > Tools > AI Assistant > MCP Servers**, click **+**, and add:
 ## CLI
 
 ```bash
-sediment           # Start MCP server
-sediment init      # Set up Claude Code integration
-sediment stats     # Show database statistics
-sediment list      # List stored items
+sediment                          # Start MCP server
+sediment init                     # Set up Claude Code integration
+sediment stats                    # Show database statistics
+sediment list                     # List stored items
+sediment list --scope global      # List global items
+sediment store "some content"     # Store content
+sediment store -                  # Store content from stdin
+sediment recall "search query"    # Search by semantic similarity
+sediment forget <id>              # Delete an item by ID
 ```
 
+All commands support `--json` for machine-readable output.
+
 ## How It Works
 
 ### Two-Database Hybrid

src/db.rs

@@ -28,7 +28,7 @@ fn sanitize_sql_string(s: &str) -> String {
 /// Validate that a string looks like a valid item/project ID (UUID hex + hyphens).
 /// Returns true if the string only contains safe characters for SQL interpolation.
 /// Use this as an additional guard before `sanitize_sql_string` for ID fields.
-pub(crate) fn is_valid_id(s: &str) -> bool {
+pub fn is_valid_id(s: &str) -> bool {
     !s.is_empty() && s.len() <= 64 && s.chars().all(|c| c.is_ascii_hexdigit() || c == '-')
 }