@bis-code/study-dash 0.5.1 → 0.6.0

package/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "study-dash",
-  "version": "0.5.1",
+  "version": "0.6.0",
   "description": "Structured learning on any subject with dashboard, Q&A logging, visualizations, exercises",
   "author": {
     "name": "Ioan-Sorin Baicoianu",

package/hooks/hooks.json

@@ -5,7 +5,7 @@
         "matcher": "learn_mark_done",
         "hooks": [{
           "type": "prompt",
-          "prompt": "The user just completed a topic. Generate 2-3 exercises for this topic using learn_create_exercise. Include at least one coding exercise and one quiz. Match the difficulty to the topic content."
+          "prompt": "The user just completed a topic. Generate 2-3 exercises for this topic using learn_create_exercise. If the subject has a programming language, include at least one coding exercise (with starter_code and test_content following the language's conventions) and one quiz. If the subject has no language, create only quiz or assignment exercises. Match the difficulty to the topic content."
         }]
       },
       {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@bis-code/study-dash",
-  "version": "0.5.1",
+  "version": "0.6.0",
   "description": "Claude Code plugin for structured learning on any subject — dashboard, Q&A logging, visualizations, exercises",
   "license": "MIT",
   "author": "Ioan-Sorin Baicoianu <baicoianuioansorin@gmail.com>",
@@ -23,7 +23,7 @@
     ".claude-plugin",
     "server/dist",
     "server/package.json",
-    "server/node_modules/better-sqlite3",
+    "server/start.sh",
     "commands",
     "skills",
     "hooks",

package/rules/tutor-mode.md

@@ -33,6 +33,18 @@ All logged content MUST be markdown:
 - Point out common mistakes and gotchas
 - Adapt language and depth to the subject domain
 
+## Exercise Generation
+
+When creating exercises for coding subjects, follow the language's conventions:
+- **Go**: `package main`, tests use `testing.T`, files are `main.go` + `main_test.go`
+- **Python**: plain functions, tests use `pytest` with `assert`, files are `main.py` + `test_main.py`
+- **Rust**: `fn` functions, tests use `#[test]` + `assert_eq!`, files are `main.rs` + `main_test.rs`
+- **TypeScript/JavaScript**: ESM exports, tests use `vitest` describe/it/expect, files are `main.ts` + `main.test.ts`
+
+For non-coding subjects (no language set): only create quiz or assignment exercises.
+
+Always provide `starter_code` and `test_content` for coding exercises.
+
 ## Dashboard
 
 The learning dashboard runs at http://127.0.0.1:19282 — mention it if the user wants to review progress.