opencode-skills-collection 4.0.74 → 4.0.76

package/bundled-skills/claude-api/typescript/claude-api/README.md

@@ -24,7 +24,7 @@ const client = new Anthropic({ apiKey: "your-api-key" });
 
 ```typescript
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   messages: [{ role: "user", content: "What is the capital of France?" }],
 });
@@ -37,7 +37,7 @@ console.log(response.content[0].text);
 
 ```typescript
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   system:
     "You are a helpful coding assistant. Always provide examples in Python.",
@@ -53,7 +53,7 @@ const response = await client.messages.create({
 
 ```typescript
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   messages: [
     {
@@ -78,7 +78,7 @@ import fs from "fs";
 const imageData = fs.readFileSync("image.png").toString("base64");
 
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   messages: [
     {
@@ -105,7 +105,7 @@ Use top-level `cache_control` to automatically cache the last cacheable block in
 
 ```typescript
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   cache_control: { type: "ephemeral" }, // auto-caches the last cacheable block
   system: "You are an expert on this large document...",
@@ -119,7 +119,7 @@ For fine-grained control, add `cache_control` to specific content blocks:
 
 ```typescript
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   system: [
     {
@@ -133,7 +133,7 @@ const response = await client.messages.create({
 
 // With explicit TTL (time-to-live)
 const response2 = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   system: [
     {
@@ -156,7 +156,7 @@ const response2 = await client.messages.create({
 ```typescript
 // Opus 4.6: adaptive thinking (recommended)
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 16000,
   thinking: { type: "adaptive" },
   output_config: { effort: "high" }, // low | medium | high | max
@@ -214,7 +214,7 @@ const messages: Anthropic.MessageParam[] = [
 ];
 
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   messages: messages,
 });
@@ -243,7 +243,7 @@ async function chat(userMessage: string): Promise<string> {
 
   const response = await client.beta.messages.create({
     betas: ["compact-2026-01-12"],
-    model: "claude-opus-4-6",
+    model: "claude-opus-5",
     max_tokens: 4096,
     messages,
     context_management: {
@@ -288,7 +288,7 @@ The `stop_reason` field in the response indicates why the model stopped generati
 ```typescript
 // Automatic caching (simplest — caches the last cacheable block)
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   cache_control: { type: "ephemeral" },
   system: largeDocumentText, // e.g., 50KB of context
@@ -303,7 +303,7 @@ const response = await client.messages.create({
 
 ```typescript
 const countResponse = await client.messages.countTokens({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   messages: messages,
   system: system,
 });

package/bundled-skills/claude-api/typescript/claude-api/batches.md

@@ -24,7 +24,7 @@ const messageBatch = await client.messages.batches.create({
     {
       custom_id: "request-1",
       params: {
-        model: "claude-opus-4-6",
+        model: "claude-opus-5",
         max_tokens: 1024,
         messages: [
           { role: "user", content: "Summarize climate change impacts" },
@@ -34,7 +34,7 @@ const messageBatch = await client.messages.batches.create({
     {
       custom_id: "request-2",
       params: {
-        model: "claude-opus-4-6",
+        model: "claude-opus-5",
         max_tokens: 1024,
         messages: [
           { role: "user", content: "Explain quantum computing basics" },

package/bundled-skills/claude-api/typescript/claude-api/files-api.md

@@ -41,7 +41,7 @@ console.log(`Size: ${uploaded.size_bytes} bytes`);
 
 ```typescript
 const response = await client.beta.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   messages: [
     {

package/bundled-skills/claude-api/typescript/claude-api/streaming.md

@@ -4,7 +4,7 @@
 
 ```typescript
 const stream = client.messages.stream({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   messages: [{ role: "user", content: "Write a story" }],
 });
@@ -27,7 +27,7 @@ for await (const event of stream) {
 
 ```typescript
 const stream = client.messages.stream({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 16000,
   thinking: { type: "adaptive" },
   messages: [{ role: "user", content: "Analyze this problem" }],
@@ -82,7 +82,7 @@ const getWeather = betaZodTool({
 });
 
 const runner = client.beta.messages.toolRunner({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 4096,
   tools: [getWeather],
   messages: [
@@ -117,7 +117,7 @@ for await (const messageStream of runner) {
 
 ```typescript
 const stream = client.messages.stream({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   messages: [{ role: "user", content: "Hello" }],
 });

package/bundled-skills/claude-api/typescript/claude-api/tool-use.md

@@ -30,7 +30,7 @@ const getWeather = betaZodTool({
 
 // The tool runner handles the agentic loop and returns the final message
 const finalMessage = await client.beta.messages.toolRunner({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 4096,
   tools: [getWeather],
   messages: [{ role: "user", content: "What's the weather in Paris?" }],
@@ -61,7 +61,7 @@ let messages: Anthropic.MessageParam[] = [{ role: "user", content: userInput }];
 
 while (true) {
   const response = await client.messages.create({
-    model: "claude-opus-4-6",
+    model: "claude-opus-5",
     max_tokens: 4096,
     tools: tools,
     messages: messages,
@@ -111,7 +111,7 @@ let messages: Anthropic.MessageParam[] = [{ role: "user", content: userInput }];
 
 while (true) {
   const stream = client.messages.stream({
-    model: "claude-opus-4-6",
+    model: "claude-opus-5",
     max_tokens: 4096,
     tools,
     messages,
@@ -169,7 +169,7 @@ while (true) {
 
 ```typescript
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   tools: tools,
   messages: [{ role: "user", content: "What's the weather in Paris?" }],
@@ -180,7 +180,7 @@ for (const block of response.content) {
     const result = await executeTool(block.name, block.input);
 
     const followup = await client.messages.create({
-      model: "claude-opus-4-6",
+      model: "claude-opus-5",
       max_tokens: 1024,
       tools: tools,
       messages: [
@@ -204,7 +204,7 @@ for (const block of response.content) {
 
 ```typescript
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   tools: tools,
   tool_choice: { type: "tool", name: "get_weather" },
@@ -224,7 +224,7 @@ import Anthropic from "@anthropic-ai/sdk";
 const client = new Anthropic();
 
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 4096,
   messages: [
     {
@@ -257,7 +257,7 @@ const uploaded = await client.beta.files.upload({
 // Code execution is GA; Files API is still beta (pass via RequestOptions)
 const response = await client.messages.create(
   {
-    model: "claude-opus-4-6",
+    model: "claude-opus-5",
     max_tokens: 4096,
     messages: [
       {
@@ -317,7 +317,7 @@ for (const block of response.content) {
 ```typescript
 // First request: set up environment
 const response1 = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 4096,
   messages: [
     {
@@ -333,7 +333,7 @@ const containerId = response1.container.id;
 
 const response2 = await client.messages.create({
   container: containerId,
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 4096,
   messages: [
     {
@@ -353,7 +353,7 @@ const response2 = await client.messages.create({
 
 ```typescript
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 2048,
   messages: [
     {
@@ -387,7 +387,7 @@ const handlers: MemoryToolHandlers = {
 const memory = betaMemoryTool(handlers);
 
 const runner = client.beta.messages.toolRunner({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 2048,
   tools: [memory],
   messages: [{ role: "user", content: "Remember my preferences" }],
@@ -424,7 +424,7 @@ const ContactInfoSchema = z.object({
 const client = new Anthropic();
 
 const response = await client.messages.parse({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   messages: [
     {
@@ -445,7 +445,7 @@ console.log(response.parsed_output.name); // "Jane Doe"
 
 ```typescript
 const response = await client.messages.create({
-  model: "claude-opus-4-6",
+  model: "claude-opus-5",
   max_tokens: 1024,
   messages: [
     {

package/bundled-skills/electron-drive-skill/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Search 3D, LLC
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/bundled-skills/electron-drive-skill/README.md

@@ -0,0 +1,172 @@
+# electron-drive-skill
+
+An agentic skill that lets an agent drive
+your Electron app the way a person would: start and stop it, read the screen,
+click, type, take screenshots, run code in the renderer or main process, and
+read logs. The agent can then check a UI change or reproduce a bug in the
+real app, not just in unit tests.
+
+|Tested On|Status|
+|--|--|
+|Claude Code|y|
+|Muse Code|y|
+
+```text
+$ drive start --build
+Using the build from 0 minute(s) ago. Pass --build to rebuild.
+{ "mode": "built", "profile": "default", "running": true, "window": { "title": "My App", ... } }
+$ drive snapshot
+- button "Get started"
+$ drive click 'role=button[name="Get started"]'
+clicked role=button[name="Get started"]
+$ drive stop
+quit via app.quit()
+```
+
+The app runs under Playwright in a background daemon, so each command is a
+separate, fast shell call. Every launch uses a scratch profile
+(`--user-data-dir` under your temp directory), so the agent never touches your
+real app data.
+
+## Requirements
+
+- Node.js 18 or later
+- `electron` and `playwright-core` 1.49 or later, installed in your project:
+  ```bash
+  npm i -D playwright-core
+  ```
+  No browsers need to be downloaded; the driver uses your project's Electron.
+- macOS or Linux. On Windows, run it as `node scripts/drive.mjs`. The
+  process-tree kill that `stop` falls back to is not implemented there.
+
+## Install
+
+Clone it into your project's skills directory:
+
+```bash
+git clone https://github.com/Search-3D/electron-drive-skill .claude/skills/electron-drive-skill
+```
+
+Or clone it into `~/.claude/skills/electron-drive-skill` to use it in every
+project. The driver always uses the Electron and Playwright of the project you
+run it in. Keep the folder name: it must match the `name` in `SKILL.md`.
+
+The examples here call it `drive`. Nothing installs that command; to run it by
+hand, add an alias:
+
+```bash
+alias drive="node .claude/skills/electron-drive-skill/scripts/drive.mjs"
+```
+
+> [!IMPORTANT]
+> Make sure you have installed playwright in your project's developer dependencies, as outlined in the steps above.
+
+## Configure
+
+With no configuration, `start` runs `electron .` on whatever is already built.
+For anything else, add `drive.config.json` at your project root (copy
+[`drive.config.json.example`](drive.config.json.example) as a starting point).
+Every field is optional:
+
+```json
+{
+  "build": "npm run build",
+  "args": ["."],
+  "env": { "APP_DATA_DIR": "{profile}/data" },
+  "logFile": "logs/main.log",
+  "dev": {
+    "args": ["."],
+    "env": { "ELECTRON_RENDERER_URL": "http://localhost:5173" },
+    "url": "http://localhost:5173"
+  }
+}
+```
+
+| Field | Meaning |
+|---|---|
+| `build` | Shell command that `start --build` runs first. |
+| `args` | Arguments to Electron: the app directory (its `package.json` `main` must be the built entry point) or the entry file. For example, electron-react-boilerplate uses `["release/app"]`. |
+| `env` | Added to the app's environment. `{profile}` becomes the scratch profile directory; it is replaced only in `env` values, not in `args`. Use it to redirect any data the app keeps outside `userData` (see below). |
+| `logFile` | A log file, relative to the profile directory, that `logs` shows along with the console output. |
+| `dev.args`, `dev.env` | Used by `start --dev` instead of `args` (plus `env`). |
+| `dev.url` | The renderer dev server. `start --dev` refuses to launch if it is not up. |
+
+### Redirecting app data with `env`
+
+Every launch gets `--user-data-dir` pointed at a scratch profile, which moves
+`app.getPath('userData')` and everything Electron stores there. Data your app
+keeps somewhere else, such as a database, a library folder or a cache path, is
+**not** moved, so the agent would work on your real files.
+
+If your app keeps data elsewhere, give it an environment variable that
+overrides the location, and set that variable in `env`:
+
+```ts
+// main process
+const dataDir = process.env.APP_DATA_DIR ?? defaultDataDir()
+```
+
+```json
+{ "env": { "APP_DATA_DIR": "{profile}/data" } }
+```
+
+`APP_DATA_DIR` is only an example name. The driver just sets the variable, so
+it has an effect only if your app reads it. Use whatever name your app
+already checks.
+
+For `--dev`, start your renderer dev server yourself, and make sure it does not
+also launch Electron.
+
+## Commands
+
+```text
+start [--dev] [--build] [--profile <name>] [--fresh]
+status                           whether the app is up, its profile and window
+snapshot                         accessibility tree of the window
+screenshot [--out <png>] [--selector <css>]
+click <target>                   Playwright selector: 'text=Settings', 'role=button[name="Save"]', css
+fill <target> <text>
+select <target> <value|label>
+press <key>                      e.g. Enter, Escape, Control+K
+wait <target> [--timeout <ms>]
+eval <js | ->                    in the renderer; '-' reads stdin
+main <js | ->                    in the main process; 'electron' and 'process' are in scope
+logs [--lines <n>]
+stop
+```
+
+Put `--` before text that starts with dashes, so it is not read as a flag:
+`drive fill 'role=textbox[name="Args"]' -- --verbose`.
+
+`eval` and `main` return their result as JSON. A DOM node, a function or
+anything else that does not serialize comes back as `undefined` or `{}`;
+return plain data instead (`el.textContent`, not `el`).
+
+Run the script directly instead of through a `yarn`/`npm` script alias. yarn 1
+re-splits arguments and drops inner quotes, which silently breaks most
+JavaScript passed to `eval` and `main`.
+
+## Limitations
+
+- **One window.** Commands act on the first window that is not DevTools.
+  There is no way to pick another one, so a settings window or a second
+  document window cannot be driven. An app with a splash screen is driven
+  once the splash closes and the main window is the only one left; check
+  `status` to see which window is current.
+- **One app at a time** per project: `start` refuses while one is running.
+- **Windows:** run it as `node scripts/drive.mjs`; the process-tree kill that
+  `stop` falls back to is POSIX only.
+
+## How it works
+
+- `scripts/drive.mjs`: the CLI. `start` spawns the daemon and waits for it.
+  Every other command is one HTTP request to it.
+- `scripts/daemon.mjs`: launches Electron through Playwright's `_electron`
+  API and serves commands on `127.0.0.1` with a random token. Its state file
+  and log are in `$TMPDIR/electron-drive/<project>-<hash>/`.
+- `scripts/lib.mjs`: config, argument parsing and paths, with no side
+  effects. Tests: `node --test scripts/`.
+
+## License
+
+[MIT](LICENSE)