@assertrx/dino 1.0.1 → 1.0.3

package/README.md

@@ -321,8 +321,9 @@ npm install -g @earendil-works/pi-coding-agent
 # 2. Install Dino via pi
 pi install npm:@assertrx/dino --approve
 
-# 3. Set API key
+# 3. Set an API key (any supported provider works)
 export ANTHROPIC_API_KEY=sk-ant-...
+# e.g. for OpenAI instead: export OPENAI_API_KEY=sk-...
 
 # 4. Start the server
 node ~/.pi/agent/npm/node_modules/@assertrx/dino/scripts/api-server.mjs
@@ -333,13 +334,35 @@ node ~/.pi/agent/npm/node_modules/@assertrx/dino/scripts/api-server.mjs
 | Method | Path | Description |
 |--------|------|-------------|
 | `GET` | `/api/health` | Health check |
+| `GET` | `/api/providers` | List all providers with auth status and model counts |
+| `GET` | `/api/models` | List models available with current credentials (`?provider=<id>` to filter) |
+| `POST` | `/api/models` | List models for given credentials — body accepts `provider`+`apiKey` or `apiKeys` (validates the key and unlocks that provider's catalog) |
 | `GET` | `/api/prompts` | List all available prompts with parameters |
 | `GET` | `/api/prompts/:name` | Get raw markdown content of a prompt |
 | `POST` | `/api/prompts/:name` | **Execute** a prompt via the pi agent |
 
+### Providers, models & API keys
+
+The server supports every pi provider (Anthropic, OpenAI, Google, DeepSeek, Groq, Mistral, xAI, OpenRouter, and many more). Credentials can come from three places, in this order of precedence:
+
+1. **Per-request keys** — pass `apiKeys` (a map of provider → key) or `apiKey` (+ optional `provider`) in the POST body. These keys are scoped to that single request's session.
+2. **Environment variables** — any provider's standard env var (e.g. `ANTHROPIC_API_KEY`, `OPENAI_API_KEY`, `GEMINI_API_KEY`).
+3. **Stored credentials** — logins created via `pi /login` in `~/.pi/agent/auth.json`.
+
+Use `GET /api/providers` to discover which providers are configured and `GET /api/models` to list models available to the server right now.
+
+**Selecting a model:** pass `model` in the POST body as either `"provider/model-id"` (e.g. `"openai/gpt-5.2"`) or `{ "provider": "openai", "id": "gpt-5.2" }`. Without `model`, the session uses the default model resolution (configured provider defaults). Requesting a model whose provider has no credentials returns a clear error.
+
 ### Prompt execution endpoints
 
-All POST bodies accept `apiKey` (optional, falls back to `ANTHROPIC_API_KEY` env var), `cwd` (workspace directory, defaults to current), and `stream` (default: `true` for SSE).
+All POST bodies accept auth/model options, `cwd` (workspace directory, defaults to current), and `stream` (default: `true` for SSE):
+
+| Body param | Description |
+|------------|-------------|
+| `apiKeys` | Map of provider → API key, e.g. `{"anthropic": "sk-...", "openai": "sk-..."}` |
+| `apiKey` | Single API key. Pair with `provider` to target one explicitly; otherwise applied to the first available provider (anthropic fallback) |
+| `provider` | Provider id (e.g. `"openai"`) to pair with a single `apiKey` |
+| `model` | Model selection: `"provider/model-id"`, `"model-id"`, or `{"provider": ..., "id": ...}` |
 
 | Prompt name | Required params |
 |-------------|---------------|
@@ -357,14 +380,33 @@ All POST bodies accept `apiKey` (optional, falls back to `ANTHROPIC_API_KEY` env
 # List all prompts
 curl http://localhost:3000/api/prompts
 
-# Get prompt details
-curl http://localhost:3000/api/prompts/setup-playwright
+# Discover providers and their auth status
+curl http://localhost:3000/api/providers
+
+# List models available to the server (optionally per provider)
+curl http://localhost:3000/api/models
+curl "http://localhost:3000/api/models?provider=anthropic"
 
-# Execute a prompt (SSE streaming)
+# Execute a prompt (SSE streaming, auth from env vars)
 curl -X POST http://localhost:3000/api/prompts/setup-playwright \
   -H "Content-Type: application/json" \
   -d '{"cwd": "/path/to/workspace"}'
 
+# Execute with a single provider key + model selection
+# (env: export OPENAI_API_KEY=sk-... then:)
+curl -X POST http://localhost:3000/api/prompts/setup-playwright \
+  -H "Content-Type: application/json" \
+  -d '{"cwd": "/path/to/workspace", "model": "openai/gpt-5.2"}'
+
+# Execute with multiple per-request API keys and an explicit model
+curl -X POST http://localhost:3000/api/prompts/setup-playwright \
+  -H "Content-Type: application/json" \
+  -d '{
+    "cwd": "/path/to/workspace",
+    "apiKeys": {"anthropic": "sk-ant-...", "openai": "sk-ov-..."},
+    "model": "anthropic/claude-sonnet-4-5"
+  }'
+
 # Execute with URL parameter (non-streaming JSON response)
 curl -X POST http://localhost:3000/api/prompts/generate-pom-from-url-cli \
   -H "Content-Type: application/json" \
@@ -399,5 +441,5 @@ event: error      → {"message": "..."}      # error
 |----------|---------|-------------|
 | `PORT` | `3000` | Server port |
 | `HOST` | `0.0.0.0` | Server host |
-| `ANTHROPIC_API_KEY` | — | API key (or pass `apiKey` in request body) |
+| Provider env vars | — | Any provider's standard env var (e.g. `ANTHROPIC_API_KEY`, `OPENAI_API_KEY`, `GEMINI_API_KEY`); per-request `apiKeys`/`apiKey` override these |
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@assertrx/dino",
-  "version": "1.0.1",
+  "version": "1.0.3",
   "description": "A friendly purple dinosaur pi package with a custom TUI header, theme, and Playwright skills.",
   "keywords": [
     "pi-package"

package/prompts/generate-pom-from-url-cli.md

@@ -1,23 +1,22 @@
 ---
-description: Generate Playwright Page Object Models from a live URL using playwright-cli
+description: Generate a Playwright Page Object Model for a single live URL using playwright-cli
 argument-hint: "<URL> [output-dir] [auth-env-file]"
 ---
 
-Use the `playwright-pom` skill and `playwright-cli` commands to generate Page Object Models from the website at $1.
+Use the `playwright-pom` skill and `playwright-cli` commands to generate a Page Object Model for the single page at $1.
 
 Scope and rules:
-- Start at $1 and explore only pages on the same origin.
+- Only generate a POM for the exact page at $1 — do not navigate to or explore any other pages.
 - Use `playwright-cli open <url>` to launch the browser. If `playwright-cli` is not on PATH, use `npx -p @playwright/cli playwright-cli open <url>` instead.
-- Use `playwright-cli snapshot` to capture the accessibility tree and element refs for each page.
-- Use `playwright-cli goto <url>` or `playwright-cli click <ref>` to navigate to additional pages.
-- Generate one focused page object per unique page or reusable component.
+- Use `playwright-cli snapshot` to capture the accessibility tree and element refs for the page. Do not use `goto` or `click` to navigate to additional pages.
+- Generate one page object for this page (plus small reusable component objects only if the page clearly contains distinct, reusable UI components).
 - Use resilient locators: `getByRole`, `getByLabel`, `getByTestId`, `getByPlaceholder`, `getByText` (in that order). Use CSS/XPath only as a last resort.
 - Each page object should expose high-level business methods (e.g., `loginAs(user)`, `searchFor(term)`) built from the low-level locators.
 - Keep selectors and URLs inside the page object; tests should read like user workflows.
 - Save generated files under `${2:-tests/pom/}` with clear, kebab-case names (e.g., `login-page.ts`).
 - Only target applications you own or have explicit authorization to test.
 - Do not interact with destructive actions (delete, remove, sign out) unless explicitly requested.
-- Stop after covering the main interactive pages; report which pages were generated and which were skipped.
+- Report what was covered once the page object is complete.
 
 ## Authentication
 
@@ -42,7 +41,7 @@ If the site requires login, handle credentials securely:
 3. **Never** ask the user to paste credentials into the chat, and **never** hardcode credentials into generated files.
 
 After generating, summarize:
-- The pages and components covered
-- The output file paths
-- Any pages that could not be inspected and why
+- The page (and any components) covered
+- The output file path(s)
+- Any parts of the page that could not be inspected and why
 - How authentication was handled (env file or manual)

package/prompts/generate-pom-from-url.md

@@ -1,21 +1,21 @@
 ---
-description: Generate Playwright Page Object Models from a live URL using the Playwright MCP bridge
+description: Generate a Playwright Page Object Model for a single live URL using the Playwright MCP bridge
 argument-hint: "<URL> [output-dir] [auth-env-file]"
 ---
 
-Use the `playwright-pom` skill and the Playwright MCP `browser_*` tools to generate Page Object Models from the website at $1.
+Use the `playwright-pom` skill and the Playwright MCP `browser_*` tools to generate a Page Object Model for the single page at $1.
 
 Scope and rules:
-- Start at $1 and explore only pages on the same origin.
-- Use `browser_navigate` to open each page and `browser_snapshot` to inspect the accessibility tree.
-- Generate one focused page object per unique page or reusable component.
+- Only generate a POM for the exact page at $1 — do not navigate to or explore any other pages.
+- Use `browser_navigate` to open the page and `browser_snapshot` to inspect the accessibility tree.
+- Generate one page object for this page (plus small reusable component objects only if the page clearly contains distinct, reusable UI components).
 - Use resilient locators: `getByRole`, `getByLabel`, `getByTestId`, `getByPlaceholder`, `getByText` (in that order). Use CSS/XPath only as a last resort.
 - Each page object should expose high-level business methods (e.g., `loginAs(user)`, `searchFor(term)`) built from the low-level locators.
 - Keep implementation details (selectors, URLs) inside the page object; tests should read like user workflows.
 - Save the generated files under `${2:-tests/pom/}` using clear, kebab-case names (e.g., `login-page.ts`).
 - Only target applications you own or have explicit authorization to test.
 - Do not interact with destructive actions (delete, remove, sign out) unless explicitly requested.
-- Stop after covering the main interactive pages; report which pages were generated and which were skipped.
+- Report what was covered once the page object is complete.
 
 ## Authentication
 
@@ -39,7 +39,7 @@ If the site requires login, handle credentials securely:
 3. **Never** ask the user to paste credentials into the chat, and **never** hardcode credentials into generated files.
 
 After generating, summarize:
-- The pages and components covered
-- The output file paths
-- Any pages that could not be inspected and why
+- The page (and any components) covered
+- The output file path(s)
+- Any parts of the page that could not be inspected and why
 - How authentication was handled (env file or manual)

package/scripts/.playwright-mcp/console-2026-08-27T18-21-56-124Z.log

@@ -0,0 +1,2 @@
+[   14570ms] [WARNING] The powerPreference option is currently ignored when calling requestAdapter() on Windows. See https://crbug.com/369219127 @ :0
+[   18833ms] [WARNING] A speculation rule set was inserted into the document but will be ignored. This might happen, for example, if it was previously inserted into another document, or if it was created using the innerHTML setter. @ https://www.google.com/search?q=playwright+test+automation&sca_esv=b3cb815f54d8ec19&source=hp&ei=RICQav7gBoDm5NoPtYioaA&iflsig=ABILxe8AAAAAapCOVCQUd3OYJje-4_I4lPG-zQWogDP_&ved=0ahUKEwj-kr3JtsGWAxUAM1kFHTUECg0Q4dUDCDM&uact=5&oq=playwright+test+automation&gs_lp=Egdnd3Mtd2l6IhpwbGF5d3JpZ2h0IHRlc3QgYXV0b21hdGlvbkgNUABYAHAAeACQAQCYAQCgAQCqAQC4AQPIAQD4AQGYAgCgAgCYAwCSBwCgBwCyBwC4BwDCBwDIBwCACAE&sclient=gws-wiz&sei=UICQauzJI6it5NoPu7KviQ0:34

package/scripts/.playwright-mcp/page-2026-08-27T18-21-56-619Z.yml

@@ -0,0 +1,52 @@
+- generic [ref=e2]:
+  - navigation [ref=e3]:
+    - link "About" [ref=e4] [cursor=pointer]:
+      - /url: https://about.google/?fg=1&utm_source=google-US&utm_medium=referral&utm_campaign=hp-header
+    - link "Store" [ref=e5] [cursor=pointer]:
+      - /url: https://store.google.com/US?utm_source=hp_header&utm_medium=google_ooo&utm_campaign=GS100042&hl=en-US
+    - generic [ref=e7]:
+      - generic [ref=e8]:
+        - link "Gmail" [ref=e10] [cursor=pointer]:
+          - /url: https://mail.google.com/mail/&ogbl
+        - link "Search for Images" [ref=e12] [cursor=pointer]:
+          - /url: https://www.google.com/imghp?hl=en&ogbl
+          - text: Images
+      - button "Google apps" [ref=e15] [cursor=pointer]
+      - link "Sign in" [ref=e20] [cursor=pointer]:
+        - /url: https://accounts.google.com/ServiceLogin?hl=en&passive=true&continue=https://www.google.com/&ec=futura_exp_og_so_72776762_e
+  - generic [ref=e22]:
+    - link [ref=e24] [cursor=pointer]:
+      - /url: /search?sca_esv=b3cb815f54d8ec19&q=Which+National+Parks+have+some+of+the+most+scenic+overlooks%3F&udm=50&oi=ddle&noiga=1&ct=460148972&peek_pws=0&aep=84&sa=X&ved=0ahUKEwj-kr3JtsGWAxUAM1kFHTUECg0QPQgH
+      - img "Celebrating National Park Week at Yosemite" [ref=e26]
+    - button "Share" [ref=e28]
+  - search [ref=e36]:
+    - generic [ref=e38]:
+      - generic [ref=e40]:
+        - button "Add files and tools" [ref=e45] [cursor=pointer]
+        - combobox "Search" [active] [ref=e50]
+        - generic [ref=e51]:
+          - generic [ref=e52]:
+            - button "Search by voice" [ref=e55] [cursor=pointer]
+            - button "Search by image" [ref=e60] [cursor=pointer]
+          - link "AI Mode" [ref=e63] [cursor=pointer]
+      - generic [ref=e76]:
+        - button "Google Search" [ref=e77] [cursor=pointer]
+        - button "I'm Feeling Lucky" [ref=e78] [cursor=pointer]
+  - 'link "Celebrate National Park Week: Explore Yosemite with Google 🌲" [ref=e88] [cursor=pointer]'
+  - contentinfo [ref=e90]:
+    - generic [ref=e91]:
+      - generic [ref=e92]:
+        - link "Advertising" [ref=e93] [cursor=pointer]:
+          - /url: https://www.google.com/intl/en_us/ads/?subid=ww-ww-et-g-awa-a-g_hpafoot1_1!o2&utm_source=google.com&utm_medium=referral&utm_campaign=google_hpafooter&fg=1
+        - link "Business" [ref=e94] [cursor=pointer]:
+          - /url: https://www.google.com/services/?subid=ww-ww-et-g-awa-a-g_hpbfoot1_1!o2&utm_source=google.com&utm_medium=referral&utm_campaign=google_hpbfooter&fg=1
+        - link "How Search works" [ref=e95] [cursor=pointer]:
+          - /url: https://google.com/search/howsearchworks/?fg=1
+      - link "Build, create, and do more with AI tools from Google" [ref=e97] [cursor=pointer]:
+        - /url: https://ai.google/helpful-tools/?utm_source=googlehpfooter&utm_medium=housepromos&utm_campaign=bottom-footer
+      - generic [ref=e98]:
+        - link "Privacy" [ref=e99] [cursor=pointer]:
+          - /url: https://policies.google.com/privacy?hl=en&fg=1
+        - link "Terms" [ref=e100] [cursor=pointer]:
+          - /url: https://policies.google.com/terms?hl=en&fg=1
+        - button "Settings" [ref=e104] [cursor=pointer]