Skip to content

Commit a7ed13c

Browse files
committed
Update CHANGELOG for 0.3.3
1 parent a20b9e7 commit a7ed13c

File tree

1 file changed

+36
-0
lines changed

1 file changed

+36
-0
lines changed

CHANGELOG.md

Lines changed: 36 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -2,6 +2,42 @@
22

33
## 2025
44

5+
### 0.3.3 (Nov 25)
6+
7+
General cleanup and fixes, and the following notable improvements:
8+
9+
#### Docs
10+
- Document `AZURE_OPENAI_ENDPOINT_EMBEDDING` (Gwyneth Peña-Siguenza).
11+
- Add link to PyBay 2025 talk video.
12+
13+
#### Core typeagent package
14+
- Simplify `load_dotenv()`.
15+
- Split embedding requests that are too large (Raphael Wirth).
16+
- Overhaul conversation metadata in storage providers.
17+
- Add extra optional keyword parameters to `create_conversation()`.
18+
- Add `Quantifier` to ingestion schema in addition to `Quantity`.
19+
- Extract knowledge concurrently (max 4 by default) (Kevin Turcios).
20+
- Fixes for `get_multiple()` implementations.
21+
- Tweak defaults in `ConversationSettings`.
22+
23+
#### MCP server
24+
- Pass LLM requests to MCP client instead of calling the OpenAI API.
25+
- Add `--database` option to MCP server.
26+
27+
#### Tools
28+
- The _tools/query.py_ tool now supports `@`-commands. Try `@help`.
29+
- Add _tools/ingest_podcast.py_ (a tool that ingests podcasts).
30+
31+
#### Testing
32+
- Fix coverage support for MCP server test.
33+
- Use an updated "Adrian Tchaikovsky" podcast data dump (Rob Gruen).
34+
- Fix Windows testing issues. Run CI on Windows too (Raphael Wirth).
35+
- Run tests in CI using secrets stored in repo (Rob Gruen).
36+
37+
#### Infrastructure
38+
- Migrate package build from _setuptools_ to _uv_build_.
39+
- Add `install-libatomic` target to `Makefile` (Bernhard Merkle).
40+
541
### 0.3.2 (Oct 22)
642

743
Brown bag release!

0 commit comments

Comments
 (0)