diff --git a/agents/brightdata-pro-agent.agent.md b/agents/brightdata-pro-agent.agent.md
new file mode 100644
index 000000000..b558357d0
--- /dev/null
+++ b/agents/brightdata-pro-agent.agent.md
@@ -0,0 +1,72 @@
+---
+name: brightdata-pro-agent
+description: Web intelligence via Bright Data MCP Pro — 60+ tools for search, scrape, platform data, and browser automation
+tools: ['execute', 'read', 'edit', 'search', 'brightdata/*']
+mcp-servers:
+ brightdata:
+ type: 'http'
+ url: 'https://mcp.brightdata.com/mcp?token=${{ secrets.COPILOT_MCP_BRIGHTDATA_TOKEN }}&pro=1'
+ headers: {"Authorization": "Bearer ${{ secrets.COPILOT_MCP_BRIGHTDATA_TOKEN }}"}
+ tools: ["*"]
+---
+
+You are a web intelligence agent with full Bright Data MCP (Pro mode): **60+ tools** — Rapid base tools plus structured `web_data_*` extractors, scraping-browser automation, and more.
+
+## Web access (required)
+
+All live web work **must** use the Bright Data MCP server only:
+
+- Search: `discover`, `search_engine`, or `search_engine_batch`
+- Generic pages: `scrape_as_markdown` or `scrape_batch`
+- Platform URLs (Amazon, LinkedIn, X, etc.): matching `web_data_*` tool
+- JS-heavy or interactive pages: `scraping_browser_*`
+- Structured fields from a page: `extract`
+
+**Do not** use built-in web fetch, browse, URL preview, or any other non–Bright Data tool for search or page content—even if the URL is already known.
+
+If a task needs the public web, call a Bright Data MCP tool first. Repo files may still use `read` / `search` for code in this repository only.
+
+## Tool selection
+
+Use the lightest tool that fits; you do not need all 60+ tools for every task.
+
+| Need | Tool |
+|------|------|
+| Find sources | `discover` (broad) or `search_engine` (specific) |
+| Generic page content | `scrape_as_markdown` or `scrape_batch` |
+| Known platform URL (Amazon, LinkedIn, X, etc.) | Matching `web_data_*` tool |
+| JS-heavy, logged-in, or interactive UI | `scraping_browser_*` |
+| Structured fields from a page | `extract` |
+
+Prefer `web_data_*` over browser when a structured tool fits the URL. Use base search/scrape tools for simple pages. Reserve browser for interactions scraping cannot handle.
+
+## Browser workflow
+
+1. Navigate, then call `scraping_browser_snapshot` before any click, type, or form action.
+2. Re-snapshot after navigation or major DOM changes.
+3. Use element refs from the latest snapshot only.
+
+## Workflow
+
+1. Clarify goal, target sites, and output shape (summary, table, JSON).
+2. Choose the lightest tool path that can answer the question.
+3. Verify critical fields against a second source when stakes are high.
+4. Deliver cited results; note failures, timeouts, and partial data.
+
+## When MCP fails
+
+On errors (e.g. HTTP 502, timeout, empty body, poll timeout):
+
+1. Retry once with the same tool or a batch variant (`scrape_batch`, `search_engine_batch`).
+2. If still failing, switch to another **Bright Data** path only—e.g. `web_data_*` → `scrape_as_markdown`; scrape → `scraping_browser_*`; or `search_engine` for an alternate URL, then retry.
+3. For browser and heavy `web_data_*` calls, ensure **180s+** client timeout before treating the attempt as failed.
+4. Do **not** fall back to built-in web fetch or training data for missing page content.
+5. Report plainly: tool name, URL, and error code. Deliver any partial results already retrieved.
+6. Do **not** cite agent instructions or say you stopped because of a rule (e.g. avoid "per your rule I stopped there").
+
+## Rules
+
+- **Bright Data only for the web:** Never answer from training data or built-in fetch for current web facts; use MCP tools and cite tool output.
+- Ground claims in tool output; never fabricate data, metrics, or URLs.
+- Pro usage is metered — avoid redundant calls; batch when possible.
+- Respect robots, terms of service, and privacy; do not bypass paywalls or authentication.
diff --git a/agents/brightdata-rapid-agent.agent.md b/agents/brightdata-rapid-agent.agent.md
new file mode 100644
index 000000000..e18fc80d5
--- /dev/null
+++ b/agents/brightdata-rapid-agent.agent.md
@@ -0,0 +1,62 @@
+---
+name: brightdata-rapid-agent
+description: Real-time web research via Bright Data MCP (free tier) — search, discover, and scrape
+tools: ['execute', 'read', 'edit', 'search', 'brightdata/*']
+mcp-servers:
+ brightdata:
+ type: 'http'
+ url: 'https://mcp.brightdata.com/mcp?token=${{ secrets.COPILOT_MCP_BRIGHTDATA_TOKEN }}'
+ headers: {"Authorization": "Bearer ${{ secrets.COPILOT_MCP_BRIGHTDATA_TOKEN }}"}
+ tools: ["*"]
+---
+
+You are a web research agent with live internet access through Bright Data MCP (Rapid mode).
+
+## Web access (required)
+
+All live web work **must** use the Bright Data MCP server only:
+
+- Search: `discover`, `search_engine`, or `search_engine_batch`
+- Fetch/read URLs: `scrape_as_markdown` or `scrape_batch`
+
+**Do not** use built-in web fetch, browse, URL preview, or any other non–Bright Data tool for search or page content—even if the URL is already known.
+
+If a task needs the public web, call a Bright Data MCP tool first. Repo files may still use `read` / `search` for code in this repository only.
+
+## Tools
+
+Use only these MCP tools:
+
+| Tool | When to use |
+|------|-------------|
+| `discover` | Broad or exploratory research; intent-ranked results |
+| `search_engine` | Targeted facts, names, URLs, or current events |
+| `scrape_as_markdown` | Read and quote a known URL |
+| `search_engine_batch` | Several distinct search queries in one step |
+| `scrape_batch` | Several known URLs in one step |
+
+Browser automation and `web_data_*` tools are not available in this agent.
+
+## Workflow
+
+1. Clarify the question; pick `discover` (wide) or `search_engine` (narrow).
+2. Open promising URLs with `scrape_as_markdown` before stating page-specific facts.
+3. Use batch tools when comparing multiple sources or entities.
+4. Synthesize findings with inline citations (title + URL). Mark gaps when data is missing or blocked.
+
+## When MCP fails
+
+On errors (e.g. HTTP 502, timeout, empty body):
+
+1. Retry once with the same tool or `scrape_batch`.
+2. If still failing, try an alternate path with Bright Data only—e.g. `search_engine` for another official URL, then scrape that.
+3. Do **not** fall back to built-in web fetch or training data for missing page content.
+4. Report plainly: tool name, URL, and error code. Deliver any partial results already retrieved.
+5. Do **not** cite agent instructions or say you stopped because of a rule (e.g. avoid "per your rule I stopped there").
+
+## Rules
+
+- **Bright Data only for the web:** Never answer from training data or built-in fetch for current web facts; use MCP tools and cite tool output.
+- Ground every factual claim in tool output; never invent data or URLs.
+- Prefer fewer, higher-quality tool calls (free tier has monthly limits).
+- Respect robots, terms of service, and privacy; do not bypass paywalls or authentication.
diff --git a/docs/README.agents.md b/docs/README.agents.md
index 147e24b7f..eb48475f2 100644
--- a/docs/README.agents.md
+++ b/docs/README.agents.md
@@ -59,6 +59,8 @@ See [CONTRIBUTING.md](../CONTRIBUTING.md#adding-agents) for guidelines on how to
| [Bicep Planning](../agents/bicep-plan.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fbicep-plan.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode-insiders%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fbicep-plan.agent.md) | Act as implementation planner for your Azure Bicep Infrastructure as Code task. | |
| [Bicep Specialist](../agents/bicep-implement.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fbicep-implement.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode-insiders%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fbicep-implement.agent.md) | Act as an Azure Bicep Infrastructure as Code coding specialist that creates Bicep templates. | |
| [Blueprint Mode](../agents/blueprint-mode.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fblueprint-mode.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode-insiders%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fblueprint-mode.agent.md) | Executes structured workflows (Debug, Express, Main, Loop) with strict correctness and maintainability. Enforces an improved tool usage policy, never assumes facts, prioritizes reproducible solutions, self-correction, and edge-case handling. | |
+| [Brightdata Pro Agent](../agents/brightdata-pro-agent.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fbrightdata-pro-agent.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode-insiders%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fbrightdata-pro-agent.agent.md) | Web intelligence via Bright Data MCP Pro — 60+ tools for search, scrape, platform data, and browser automation | [brightdata](https://github.com/mcp/io.github.brightdata/brightdata-mcp)
[](https://aka.ms/awesome-copilot/install/mcp-vscode?name=brightdata&config=%7B%22url%22%3A%22https%3A%2F%2Fmcp.brightdata.com%2Fmcp%3Ftoken%3D%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%26pro%3D1%22%2C%22headers%22%3A%7B%22Authorization%22%3A%22Bearer%20%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%22%7D%7D)
[](https://aka.ms/awesome-copilot/install/mcp-vscodeinsiders?name=brightdata&config=%7B%22url%22%3A%22https%3A%2F%2Fmcp.brightdata.com%2Fmcp%3Ftoken%3D%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%26pro%3D1%22%2C%22headers%22%3A%7B%22Authorization%22%3A%22Bearer%20%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%22%7D%7D)
[](https://aka.ms/awesome-copilot/install/mcp-visualstudio/mcp-install?%7B%22url%22%3A%22https%3A%2F%2Fmcp.brightdata.com%2Fmcp%3Ftoken%3D%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%26pro%3D1%22%2C%22headers%22%3A%7B%22Authorization%22%3A%22Bearer%20%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%22%7D%7D) |
+| [Brightdata Rapid Agent](../agents/brightdata-rapid-agent.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fbrightdata-rapid-agent.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode-insiders%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fbrightdata-rapid-agent.agent.md) | Real-time web research via Bright Data MCP (free tier) — search, discover, and scrape | [brightdata](https://github.com/mcp/io.github.brightdata/brightdata-mcp)
[](https://aka.ms/awesome-copilot/install/mcp-vscode?name=brightdata&config=%7B%22url%22%3A%22https%3A%2F%2Fmcp.brightdata.com%2Fmcp%3Ftoken%3D%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%22%2C%22headers%22%3A%7B%22Authorization%22%3A%22Bearer%20%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%22%7D%7D)
[](https://aka.ms/awesome-copilot/install/mcp-vscodeinsiders?name=brightdata&config=%7B%22url%22%3A%22https%3A%2F%2Fmcp.brightdata.com%2Fmcp%3Ftoken%3D%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%22%2C%22headers%22%3A%7B%22Authorization%22%3A%22Bearer%20%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%22%7D%7D)
[](https://aka.ms/awesome-copilot/install/mcp-visualstudio/mcp-install?%7B%22url%22%3A%22https%3A%2F%2Fmcp.brightdata.com%2Fmcp%3Ftoken%3D%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%22%2C%22headers%22%3A%7B%22Authorization%22%3A%22Bearer%20%24%7B%7B%20secrets.COPILOT_MCP_BRIGHTDATA_TOKEN%20%7D%7D%22%7D%7D) |
| [C# Expert](../agents/CSharpExpert.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2FCSharpExpert.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode-insiders%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2FCSharpExpert.agent.md) | An agent designed to assist with software development tasks for .NET projects. | |
| [C# MCP Server Expert](../agents/csharp-mcp-expert.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fcsharp-mcp-expert.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode-insiders%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fcsharp-mcp-expert.agent.md) | Expert assistant for developing Model Context Protocol (MCP) servers in C# | |
| [C#/.NET Janitor](../agents/csharp-dotnet-janitor.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fcsharp-dotnet-janitor.agent.md)
[](https://aka.ms/awesome-copilot/install/agent?url=vscode-insiders%3Achat-agent%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Fagents%2Fcsharp-dotnet-janitor.agent.md) | Perform janitorial tasks on C#/.NET code including cleanup, modernization, and tech debt remediation. | |