Highlights
Trove is now a two-layer system: a Rust repository memory engine plus the existing Claude Code plugin.
- Repository indexing - symbols, module summaries, subsystem clusters, architecture map (
.trove/) - Progressive retrieval - architecture → subsystem → module → symbol, with context budgets
- Memory hierarchy - L0–L6 model documented in VISION.md
- Plugin extended -
/trove:index,/trove:query, SessionEnd reflection hook
Engine (new)
cargo build -p trove-cli
cargo run -p trove-cli -- index
cargo run -p trove-cli -- status
cargo run -p trove-cli -- query "modify vendor onboarding"CLI commands: index, status, query, import-historical, record, patch
Languages: Rust, Python, JavaScript/TypeScript, Go, Bash
Plugin (v0.1 → v0.2)
| New | Description |
|---|---|
/trove:index |
Index the current repo |
/trove:query |
Progressive retrieval before editing |
| SessionEnd hook | Reflection + re-index reminder |
Existing commands unchanged: /trove:init, /trove:remember, /trove:recall.
Breaking changes
None for existing plugin users. Engine features require Rust (cargo build -p trove-cli).
Known limitations
- Module summaries are heuristic (not LLM-generated)
- Semantic layer uses keyword overlap (embeddings slot ready)
- Dependency graph covers imports, not full call graphs
- No pre-built binaries yet (build from source)
Upgrade
Plugin: /reload-plugins
Engine:
git pull && cargo build -p trove-cli && cargo run -p trove-cli -- index