Skip to content

Release 0.27.0-beta.1 #2069

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Aaronontheweb merged 1 commit into from
Aug 27, 2026
Merged

Release 0.27.0-beta.1 #2069

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 3 additions & 2 deletions Directory.Build.props
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,8 +8,9 @@
<LangVersion>latest</LangVersion>
<Nullable>enable</Nullable>
<TreatWarningsAsErrors>true</TreatWarningsAsErrors>
<VersionPrefix>0.26.0</VersionPrefix>
<PackageReleaseNotes>Netclaw v0.26.0 — Less approval fatigue, more token efficiency, MCP reliability + OAuth improvements, and MCP server prompts invocable as skills. Agents reach for the right tool (tool calls cut more than half on attachment tasks), structured tools are selected in 25/25 eval trials, native tool names can no longer leak into the shell, OAuth refresh survives cold restarts, and server prompts plug into Netclaw's skill system. Now proven by an executable behavioral eval harness, not vibes.</PackageReleaseNotes>
<VersionPrefix>0.27.0</VersionPrefix>
<VersionSuffix>beta.1</VersionSuffix>
<PackageReleaseNotes>Netclaw v0.27.0-beta.1 — Local semantic memory. NetClaw now embeds, recalls, and relevance-gates cross-session memories entirely on-device with ONNX — no cloud embeddings. Pinned int8 models cut steady-state RSS by 57% and run ~1.7x faster than fp32, `netclaw memory backfill-embeddings` embeds an existing corpus, and `netclaw doctor` watches embedding model health.</PackageReleaseNotes>
</PropertyGroup>
<PropertyGroup>
<NetCoreTestVersion>net10.0</NetCoreTestVersion>
Expand Down
21 changes: 21 additions & 0 deletions RELEASE_NOTES.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,26 @@
# NetClaw Release Notes

## 0.27.0-beta.1 (2026-08-27)

This beta opens the semantic memory cycle. NetClaw now embeds and recalls memories locally on your machine with ONNX — no cloud embeddings, no external services. Recall gets a relevance gate that keeps weak matches out, and `netclaw doctor` watches your embedding model's health.

### Local semantic memory (ONNX)

A new `Netclaw.Embeddings` assembly gives every cross-session memory a local embedding, searched and gated on-device.

- **Embed at write time, recall by meaning.** Memories are embedded when stored, nominated by kNN during curation, and found through hybrid vector + lexical recall ([#1749](https://github.com/netclaw-dev/netclaw/pull/1749)).
- **A cross-encoder keeps weak matches out.** `OnnxCrossEncoderScorer` applies a post-floor relevance gate so vague hits don't slip into recall ([#1749](https://github.com/netclaw-dev/netclaw/pull/1749)).
- **Local, pinned models.** Defaults are `snowflake-arctic-embed-m-int8` for embedding and `ms-marco-minilm-l-6-v2` for the gate. Both come from a pinned allowlist — NetClaw never loads an arbitrary model ([#1749](https://github.com/netclaw-dev/netclaw/pull/1749)).
- **Faster and lighter at rest.** The int8 embedder used about 57% less steady-state RSS than fp32 and ran about 1.7 times faster ([#1749](https://github.com/netclaw-dev/netclaw/pull/1749)).
- **Provisioning you can control.** Models download on first daemon start, alert through the operational channel on failure, and can be pre-provisioned for a fully offline start ([#1749](https://github.com/netclaw-dev/netclaw/pull/1749)).
- **New CLI and health tooling.** `netclaw memory backfill-embeddings` embeds an existing corpus, and `netclaw doctor` reports embedding model health ([#1749](https://github.com/netclaw-dev/netclaw/pull/1749)).

### Small fixes

- Fixed release-note parsing so version metadata resolves cleanly ([#2067](https://github.com/netclaw-dev/netclaw/pull/2067)).

> **Upgrade note:** embeddings default to enabled. Expect roughly 800 MB total RSS on the reference configuration, and plan for model access on first start. Operators can disable embeddings or pre-provision the models for offline use.

## 0.26.0 (2026-08-26)

This release closes the 0.26 beta cycle. Across five betas and a post-beta polish pass, NetClaw got faster with your permission, lighter on your token budget, and more dependable with MCP servers.
Expand Down
Loading