Dev

Cargo.toml

@@ -17,7 +17,7 @@ members = [
 resolver = "2"
 
 [workspace.package]
-version = "0.1.13"
+version = "0.1.14"
 description = "Knowing by reasoning, not vectors."
 edition = "2024"
 authors = ["zTgx <beautifularea@gmail.com>"]

HISTORY.md

@@ -1,5 +1,24 @@
 # HISTORY
 
+## 0.1.14 (2026-04-29)
+
+- Homepage redesign with product showcase and improved UX
+- Chain command and intent-based navigation hints for agent navigation
+- Parser registry and custom format support
+- Standardized import order and code formatting across modules
+- Updated package metadata and documentation links
+- Single-document reasoning challenge example with README
+- Removed redundant documentation links and unused test
+
+## 0.1.13 (2026-04-26)
+
+- **Compiler terminology**: renamed `index` to `compile`, `reindexing` to `recompilation`
+- Removed unused incremental updater and summary strategies
+- Moved keyword extraction from scoring to utils module
+- Added tracing and logging for better debugging experience
+- Workspace restructuring: renamed `vectorless-core` to `crates`
+- Removed commented-out strategy layer crates from Cargo.toml
+
 ## 0.1.12 (2026-04-24)
 
 - **Compile pipeline**: renamed index pipeline to compile pipeline with passes-based architecture

examples/vectorless/README.md

@@ -0,0 +1,51 @@
+# Single-Document Reasoning Challenge
+
+A demonstration of Vectorless's ability to perform deep reasoning on complex technical documents.
+
+## Overview
+
+This project compiles a realistic quantum computing research report and asks questions that require:
+- Cross-referencing information across distant sections
+- Tracing dependency chains between different entities
+- Extracting details buried in nested structures
+- Multi-step reasoning beyond simple keyword matching
+
+## Installation
+
+Requires the Vectorless Python SDK:
+
+```bash
+pip install vectorless
+```
+
+## Usage
+
+Set your LLM API credentials and run:
+
+```bash
+LLM_API_KEY=sk-xxx LLM_MODEL=gpt-4o LLM_ENDPOINT=https://api.openai.com/v1 python main.py
+```
+
+### Environment Variables
+
+| Variable | Description | Default |
+|----------|-------------|---------|
+| `LLM_API_KEY` | Your LLM provider API key | `sk-...` |
+| `LLM_MODEL` | Model identifier | `gpt-4o` |
+| `LLM_ENDPOINT` | API endpoint URL | `https://api.openai.com/v1` |
+
+## Challenge Questions
+
+1. **Refrigerator cost & location** — Connects Lab B's characterization requirements with Lab A's equipment specs and capital expenditure data
+
+2. **Materials dependency** — Traces how Lab C's error correction milestone depends on Lab A's materials science improvement
+
+3. **Firmware bug impact** — Calculates affected qubits by connecting Lab D's incident report with Lab A's hardware configuration
+
+4. **Gap to target** — Computes the difference between current achievement and future goals using derived values
+
+5. **Revenue coverage** — Evaluates whether a single revenue source can cover projected capital needs
+
+## License
+
+Apache-2.0

examples/single_doc_challenge.py → examples/vectorless/main.py

@@ -7,12 +7,6 @@
 the engine to navigate deep into the tree, cross-reference details
 across distant sections, and extract information buried in nested
 structures — not surface-level keyword matches.
-
-```bash
-LLM_API_KEY=sk-xxx LLM_MODEL=gpt-4o \
-  LLM_ENDPOINT=https://api.openai.com/v1 \
-  python examples/single_doc_challenge.py
-```
 """
 
 import asyncio