@dreb/coding-agent 2.25.2 → 2.25.3
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +2 -0
- package/README.md +1 -1
- package/docs/providers.md +7 -0
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -52,6 +52,8 @@
|
|
|
52
52
|
- Fixed RPC `get_session_stats` to expose `contextUsage`, so headless clients can read actual current context-window usage instead of deriving it from token totals ([#2550](https://github.com/badlogic/pi-mono/issues/2550))
|
|
53
53
|
- Fixed `pi update` for git packages to fetch only the tracked target branch with `--no-tags`, reducing unrelated branch and tag noise while preserving force-push-safe updates ([#2548](https://github.com/badlogic/pi-mono/issues/2548))
|
|
54
54
|
- Fixed print and JSON modes to emit `session_shutdown` before exit, so extensions can release long-lived resources and non-interactive runs terminate cleanly ([#2576](https://github.com/badlogic/pi-mono/issues/2576))
|
|
55
|
+
- Fixed Kimi For Coding OAuth provider to accept OpenAI-style multimodal content arrays with base64 `image_url` data URLs for `kimi-for-coding`, and ensured `modifyModels()` preserves image capability even when static metadata is stale. ([#245](https://github.com/aebrer/dreb/issues/245))
|
|
56
|
+
- Fixed silent image dropping in the `openai-completions` provider — text-only models now receive explicit placeholder text (`[image omitted: model does not support images]`) instead of silently stripping image content from user messages and tool results. ([#245](https://github.com/aebrer/dreb/issues/245))
|
|
55
57
|
|
|
56
58
|
## [0.62.0] - 2026-03-23
|
|
57
59
|
|
package/README.md
CHANGED
|
@@ -117,7 +117,7 @@ For each built-in provider, dreb maintains a list of tool-capable models, update
|
|
|
117
117
|
- MiniMax
|
|
118
118
|
- MiniMax (China)
|
|
119
119
|
|
|
120
|
-
See [docs/providers.md](docs/providers.md) for detailed setup instructions.
|
|
120
|
+
See [docs/providers.md](docs/providers.md) for detailed setup instructions, including Kimi For Coding notes that distinguish OAuth, API-key, first-party CLI, and Moonshot Open Platform vision support.
|
|
121
121
|
|
|
122
122
|
**Custom providers & models:** Add providers via `~/.dreb/agent/models.json` if they speak a supported API (OpenAI, Anthropic, Google). For custom APIs or OAuth, use extensions. See [docs/models.md](docs/models.md) and [docs/custom-provider.md](docs/custom-provider.md).
|
|
123
123
|
|
package/docs/providers.md
CHANGED
|
@@ -40,6 +40,13 @@ Use `/logout` to clear credentials. Tokens are stored in `~/.dreb/agent/auth.jso
|
|
|
40
40
|
- Requires ChatGPT Plus or Pro subscription
|
|
41
41
|
- Personal use only; for production, use the OpenAI Platform API
|
|
42
42
|
|
|
43
|
+
### Kimi For Coding
|
|
44
|
+
|
|
45
|
+
- `/login` uses the Kimi Code OAuth subscription endpoint at `https://api.kimi.com/coding/v1`.
|
|
46
|
+
- `KIMI_API_KEY` uses Kimi For Coding's Anthropic-compatible API at `https://api.kimi.com/coding`.
|
|
47
|
+
- The OAuth coding endpoint has been verified to accept OpenAI-style multimodal content arrays with base64 `image_url` data URLs for `kimi-for-coding`.
|
|
48
|
+
- Moonshot Open Platform vision docs use a different base URL (`https://api.moonshot.ai/v1`), and the first-party Kimi CLI may have its own media handling; don't assume all Kimi routes expose identical media behavior.
|
|
49
|
+
|
|
43
50
|
## API Keys
|
|
44
51
|
|
|
45
52
|
### Environment Variables or Auth File
|