xonecas.io
64875
Mc

mini-coder / commit fddb3af

chore: update Vercel AI SDK dependencies

xonecas2026-03-13 fddb3af6baf5a38a711e80c2003d176c98c39c89 parent d3862f1

2 files changed

TODO.md+4−2
@@ -1,7 +1,9 @@
11 # TODO
22
3-- ESC doesn't interrupt subagents created by custom commands, example: can't interrupt a review agent
4-- Surface to the user when compacting happens in a structured output
3+- Surface to the user when context prunning happens in a structured output
4+- Normalize the different models reasoning output. Trim excessive whitespace safely, ensure it renders uniquely. Should have a more structured output so it's noticeable that is reasoning and not a normal response. Enable the markdown rendering for it.
5+- Improve our autocomplete via TAB. it should work on all commands and parameters. When the user uses tab assume it's a file path autocompletion.
6+- Tool hooks output prints inline with spinner
57
68 ---
79
package.json+5−5
@@ -19,12 +19,12 @@
1919 "jscpd": "jscpd src"
2020 },
2121 "dependencies": {
22- "@ai-sdk/anthropic": "^3.0.58",
23- "@ai-sdk/google": "^3.0.43",
24- "@ai-sdk/openai": "^3.0.41",
25- "@ai-sdk/openai-compatible": "^2.0.35",
22+ "@ai-sdk/anthropic": "^3.0.60",
23+ "@ai-sdk/google": "^3.0.51",
24+ "@ai-sdk/openai": "^3.0.45",
25+ "@ai-sdk/openai-compatible": "^2.0.36",
2626 "@modelcontextprotocol/sdk": "^1.27.1",
27- "ai": "^6.0.116",
27+ "ai": "^6.0.127",
2828 "ignore": "^7.0.5",
2929 "yoctocolors": "^2.1.2",
3030 "zod": "^4.3.6"