Skip to content

Commit e064cf0

Browse files
taltastaltas
authored
chore: prepare v3.78.0 release (#1249)
1 parent 732651f commit e064cf0

41 files changed

Lines changed: 4066 additions & 172 deletions

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

‎CHANGELOG.md‎

Lines changed: 42 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,47 @@
11
# Zoo Code Changelog
22

3+
## [3.78.0]
4+
5+
### Minor Changes
6+
7+
- Add NanoGPT as a configurable provider with dynamic model discovery, streaming and prompt completions, and routing preferences for speed, price, latency, throughput, tool support, and caching (PR #1239 by @taltas)
8+
- Add the new Gemini 3.7 Flash model to Google Gemini and Vertex AI with a 1M context window, multimodal input, prompt caching, and configurable reasoning (PR #1241 by @app/zoomote)
9+
- Add the new GLM 5.3 model to Z AI coding plans and OpenCode Go with a 1M context window, prompt caching, and extended reasoning controls (PR #1244 by @app/zoomote)
10+
- Add the new Qwen3.8 Max model to OpenCode Go with multimodal input, caching, streamed reasoning, and Anthropic Messages routing (PR #1245 by @app/zoomote)
11+
- Fix Azure OpenAI resource endpoints and improve Azure-specific setup guidance in OpenAI Compatible settings (#1191 by @edelauna, PR #1192 by @app/zoomote)
12+
- Preserve task-history titles when rapidly navigating away before a task's messages finish loading (#1180 by @edelauna, PR #1181 by @edelauna)
13+
- Correct Kimi Code output-token defaults and honor model limits returned by the server (#1215 by @myk1yt, PR #1217 by @myk1yt)
14+
- Update DeepSeek V4 Pro reasoning efforts and normalize medium, high, and extended reasoning mappings (#1235 by @WHMHammer, PR #1236 by @WHMHammer)
15+
- Correct DeepSeek V4 pricing and expand provider coverage for the V4 Pro 0813 checkpoint (PR #1237 by @app/zoomote)
16+
- Rename the default settings import/export file to `zoo-code-settings.json` throughout the extension (#1176 by @Rafael-Silva-Oliveira, PR #1177 by @Rafael-Silva-Oliveira)
17+
- Add Destructive Command Guard support for Intel-based macOS systems (PR #1213 by @app/zoomote)
18+
- Update `undici` to 6.28.0 to address security vulnerabilities (PR #1161 by @app/renovate)
19+
- Update Mermaid to 11.16.1 to address a prototype-pollution vulnerability (PR #1193 by @app/renovate)
20+
- Record tool usage centrally to prevent duplicate telemetry and sanitize raw tool names (PR #1073 by @edelauna)
21+
- Canonicalize shared provider settings identifiers and add registry-alignment coverage (PR #1109 by @WebMad)
22+
- Canonicalize provider identifiers across CLI configuration, environment mappings, and model selection (PR #1110 by @WebMad)
23+
- Complete the webview migration to canonical provider identifiers across provider settings and routing (PR #1141 by @WebMad)
24+
- Use canonical provider identifiers throughout API options and add focused interaction coverage (PR #1146 by @WebMad)
25+
- Canonicalize provider model configuration identifiers and provider-specific settings behavior (PR #1147 by @WebMad)
26+
- Migrate model-selection UI hooks to canonical provider identifiers (PR #1148 by @WebMad)
27+
- Reuse the retired Roo provider identifier registry while preserving migration compatibility (PR #1166 by @WebMad)
28+
- Introduce typed shared test utilities for API options, filesystem mocks, reset operations, VS Code doubles, and webview rendering (PR #1171 by @app/zoomote)
29+
- Reuse shared API option factories across Requesty, OpenRouter, and Vercel AI Gateway provider tests (PR #1178 by @app/zoomote)
30+
- Reuse the shared Responses client mock in X.AI provider tests (PR #1182 by @app/zoomote)
31+
- Reuse shared VS Code context, URI, and reset helpers in custom-mode configuration tests (PR #1190 by @app/zoomote)
32+
- Reuse shared reset helpers across code-index embedder tests (PR #1194 by @app/zoomote)
33+
- Reuse shared config test helpers and remove obsolete lint suppressions (PR #1195 by @app/zoomote)
34+
- Reuse shared webview render helpers across focused chat tests (PR #1196 by @app/zoomote)
35+
- Reuse shared reset helpers across terminal integration tests (PR #1197 by @app/zoomote)
36+
- Reuse shared VS Code and reset helpers in settings import/export tests (PR #1198 by @app/zoomote)
37+
- Reuse shared reset helpers across additional code-index tests (PR #1199 by @app/zoomote)
38+
- Complete another batch of shared reset-helper adoption in code-index tests (PR #1200 by @app/zoomote)
39+
- Finish reset-helper adoption in Semble and terminal test suites (PR #1201 by @app/zoomote)
40+
- Complete shared reset-helper adoption across provider tests (PR #1202 by @app/zoomote)
41+
- Reuse shared webview render helpers across chat and settings tests (PR #1203 by @app/zoomote)
42+
- Complete shared webview render-helper adoption in the remaining settings tests (PR #1204 by @app/zoomote)
43+
- Merge the v3.76.0 release preparation branch into `main` (PR #1173 by @navedmerchant)
44+
345
## [3.76.0]
446

547
### Minor Changes

‎README.md‎

Lines changed: 6 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -46,13 +46,13 @@ Zoo Code builds on the foundation created by Roo Code and continues to expand it
4646
- **More dependable terminal and editing workflows** — fixes for premature terminal completion, task-state races, context management, diff editing, and provider-specific tool use.
4747
- **More control over your workspace** — rules management, per-mode MCP restrictions, multi-root path controls, model reasoning options, and completion change review actions.
4848

49-
## What's New in v3.76.0
49+
## What's New in v3.78.0
5050

51-
- **Run longer, uninterrupted tasks with Destructive Command Guard (DCG)** — DCG blocks dangerous commands while letting Zoo keep working without you continuously pressing approval buttons, backed by hardened managed-binary downloads and installation.
52-
- **Better provider controls and reliability** — choose OpenAI Codex response speed, use updated DeepSeek configurations, and benefit from stronger isolation between provider-profile changes and running tasks.
53-
- **Critical terminal execution fix** — Zoo now waits for terminal commands to finish before starting the next step, preventing overlapping work and premature model continuation.
54-
- Smarter batching groups related tool approvals while keeping unrelated requests separate.
55-
- Telemetry delivery and model-cache fetching are more resilient under failures and concurrent requests.
51+
- **Three major new models have arrived** — use the brand-new Gemini 3.7 Flash, GLM 5.3, and Qwen3.8 Max models, plus updated DeepSeek V4 reasoning, pricing, and provider coverage.
52+
- **Connect to NanoGPT** — use dynamic model discovery, streaming and prompt completions, and routing preferences for speed, price, latency, throughput, tool support, and caching.
53+
- **More reliable providers and tasks** — fixes improve Azure OpenAI endpoint setup, Kimi Code output limits, task-history title preservation, and Zoo settings import/export.
54+
- Destructive Command Guard now supports Intel-based Macs.
55+
- Security updates address vulnerabilities in `undici` and Mermaid.
5656

5757
<details>
5858
<summary>🌐 Available languages</summary>

‎locales/ca/README.md‎

Lines changed: 6 additions & 6 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

‎locales/de/README.md‎

Lines changed: 6 additions & 6 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

‎locales/es/README.md‎

Lines changed: 6 additions & 6 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

‎locales/fr/README.md‎

Lines changed: 6 additions & 6 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

0 commit comments

Comments
 (0)