Screenshots of live websites. Minimal interface, powerful engine.
Add setup prompt for capture tool in README
Bridger Towerc4db581a9873
diff --git a/README.md b/README.mdindex 79c9672..7e56459 100644--- a/README.md+++ b/README.md@@ -55,6 +55,8 @@ Or use the equivalent configuration in another MCP client:}```+## Setup Prompt+The server exposes one tool, `capture`, which returns the image inline with structured metadata. It writes nothing by default; pass `output` when the agent also needs a file.```json@@ -74,8 +76,7 @@ Bare localhost, `.localhost`, and loopback addresses use HTTP automatically. OthMCP clients may need a fresh agent task before the newly registered `capture` tool appears.-<details>-<summary>Give this setup prompt to your coding agent</summary>+Give this setup prompt to your coding agent</summary>```textSet up Iris as your visual camera in this coding environment.@@ -119,8 +120,6 @@ Do not claim MCP success from the CLI test alone. Do not add browser automation,interaction scripting, or review tooling; use Iris only as the camera.```-</details>-## What it does for you- Renders with your real installed Chrome, driven over the DevTools Protocol