Skip to content

CHANGELOG.md

Latest commit

 

History

History
46 lines (34 loc) · 3.2 KB

CHANGELOG.md

File metadata and controls

46 lines (34 loc) · 3.2 KB
Raw

Changelog

All notable changes to resonant-archive will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

Unreleased

0.1.0 - 2026-04-09

Initial public release. Modernized from an earlier internal tool (vault-archive-product) that was built, shelved during a cloud migration, and revived for this release.

Added

  • Python 3.11/3.12 package with pyproject.toml, setuptools backend, and a resonant-archive console script entry point
  • resonant-archive import — chunks and embeds a directory of markdown or text files into a local semantic archive, tagged with a user-supplied namespace
  • resonant-archive search — command-line semantic search with optional namespace filtering and rich table output
  • resonant-archive stats — per-namespace chunk counts and archive location
  • resonant-archive serve — FastAPI daemon that keeps the embedding model warm for fast MCP queries
  • resonant-archive mcp — stdio MCP server (for Claude Desktop) forwarding tool calls to the daemon over HTTP
  • Frontmatter-aware chunker with two strategies:
    • paragraph (default) — header-aware greedy packing with tail merging
    • window — fixed-size character windows with overlap and no-gap guarantees
  • Tiered metadata model — every chunk carries stable chunk_id, source_file, chunk_index, and char_offset; plus optional title, timestamp, tags, and Nexus-style conversation_id / provider when extractable from frontmatter
  • Namespace tagging — every chunk is stored with a namespace metadata field; results surface the namespace so querying AIs know where a hit came from
  • MCP tools — archive_search(query, namespace?, n_results?) and archive_stats(), both exposed via stdio MCP server
  • Skill file (skills/resonant-archive.md) — drops into any MCP-compatible AI's context to enable natural-language operation
  • Comprehensive docs — README, START_HERE, SETUP_GUIDE, USAGE_GUIDE, QUICK_REFERENCE, TROUBLESHOOTING, CLOUD_IMPORT
  • Cross-platform launcher scripts — START_WINDOWS.bat and START_MAC_LINUX.sh for double-click daemon startup
  • 30 unit tests covering the chunker (19) and store (11), including semantic similarity verification and namespace isolation

Technical choices

  • Embedding model: sentence-transformers/all-MiniLM-L6-v2 (384-dim, ~90 MB, fast). Alternatives documented in README.
  • Vector store: chromadb.PersistentClient — embedded, file-based, single process, no separate server
  • MCP framework: Official Python mcp SDK
  • CLI framework: typer with rich for progress bars and tables
  • Frontmatter parser: python-frontmatter (handles YAML frontmatter with graceful fallback)

Licensed under

Codependent AI Source-Available License — free for personal, educational, and non-commercial use with attribution. Commercial use requires a license.