summaryrefslogtreecommitdiff
path: root/SCRATCHPAD.md
diff options
context:
space:
mode:
authorPaul Buetow <paul@buetow.org>2025-09-15 09:22:05 +0300
committerPaul Buetow <paul@buetow.org>2025-09-15 09:22:05 +0300
commitf645911896634752d55fad50d52365c0255bb279 (patch)
tree6b40fa2e85ee1b2d187e2127758a46d7dcd971e6 /SCRATCHPAD.md
parented91ea50b56fa6b0d4cc6d80ab7cc6d3584bb3d1 (diff)
update docs
Diffstat (limited to 'SCRATCHPAD.md')
-rw-r--r--SCRATCHPAD.md34
1 files changed, 34 insertions, 0 deletions
diff --git a/SCRATCHPAD.md b/SCRATCHPAD.md
new file mode 100644
index 0000000..afea5aa
--- /dev/null
+++ b/SCRATCHPAD.md
@@ -0,0 +1,34 @@
+# Project scratch pad
+
+This document shows future items and items in progress. Already completed ones are deleted from this document as updates occur.
+
+## Features
+
+* [ ] Keep global stats about LLM usage for the tmux pane!
+* [ ] No a feature, but verify my OpenAI API account so I can use GPT-5 via the API.
+* [ ] In-editor chat triggers should be context aware of the current file, buffer and function!
+* [ ] Kagi FastGPT for in-editor search
+ - Think about an in-editor chat trigger, maybe with S> for search!
+* [ ] Test whethe GitHub Copilot support actually works now, and if not, fix it!
+* [ ] Be able to re-configure the temperature in-editor
+* [ ] For in-editor chat add a way to print current hexai status such as
+ * active LLM
+ * active stats
+ * current config printed out
+ * could use a special keyword like /status?TRIGGER (e.g. > as TRIGGER)
+ * what other slasm commands could we think of?
+* [ ] Be able to switch LLMs ad-hoc.
+
+## More
+
+* [ ] Configure hexai-lsp multiple times, but with different LLM backends? E.g. a remote cloud one and a local one?
+* [/] Review documentation
+* [/] Manual review the code
+* [ ] Useful: https://deepwiki.com/helix-editor/helix/4.3-language-server-protocol
+
+## Additional Ideas (Nice to Have)
+
+* **LSP: Inlay Hints (`textDocument/inlayHint`)**: Use AI to provide dynamic, inline hints like inferred types, performance warnings (e.g., `O(n^2)`), or security alerts.
+* **LSP: Hover (`textDocument/hover`)**: Enrich hover popups with AI-generated natural language explanations of code, usage examples, and complexity analysis.
+* **LSP: Semantic Tokens (`textDocument/semanticTokens`)**: Implement semantic tokenization to build a deeper understanding of the code, which would improve the accuracy and context-awareness of all other AI features.
+* **LSP: Rename (`textDocument/rename`)**: Add safe rename capabilities, potentially enhanced with AI to proactively suggest better, more descriptive names for variables and functions.