# browse_discover Tier-0 front door for the current session page (or pass url): does the site offer an agent-native interface (llms.txt / OpenAPI / ai-plugin)? Prefer it over scraping. Agent View of the PolicyLayer registry record for `browse_discover`. HTML page: https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-discover ## Facts - Tool: `browse_discover` - Server: Gateway (`https://wingmanprotocol.com/mcp`) — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway.md - Homepage: https://github.com/WingmanProtocol-Agent-Gateway/wingman-agent-gateway - Risk category: Read (Low risk) - Registry record: grade F, identity unverified - Server auth posture: open - Server rate-limited: no - Parameters: 2 (1 required) - Recommended policy verdict: Allowed ## Parameters | Parameter | Type | Required | Description | | --- | --- | --- | --- | | `url` | string | no | optional: probe this url instead of the current page | | `browser_id` | string | yes | from browse_open | Parameters from the server's own tool schema. ## Example call (MCP tools/call, JSON-RPC 2.0) ```json { "jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": { "name": "browse_discover", "arguments": { "browser_id": "" } } } ``` ## Why browse_discover is rated Low browse_discover performs information retrieval only: it checks whether a given URL or current session page exposes agent-native interface metadata (llms.txt, OpenAPI, ai-plugin). This is a read operation with no side effects, no command execution, and no data modification. The low severity reflects that misuse would only result in querying public interface metadata, with minimal blast radius. From the tool's own definition: "Tool description states it 'does the site offer an agent-native interface' — purely a discovery/detection function that queries and retrieves information about a website's available interfaces. No data is modified, deleted, or executed." Risk signals: Accepts URL/endpoint input (url) ## Use case AI agents call browse_discover to retrieve information from Gateway without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere. ## Recommended policy (PolicyLayer) Verdict: **Allowed**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches Gateway: ```json { "version": "1", "default": "deny", "tools": { "browse_discover": {} } } ``` ## Other tools on Gateway (270) - `cancel_watch` — Destructive — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/cancel-watch.md - `forget_memories` — Destructive — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/forget-memories.md - `text_truncate_ellipsis` — Destructive — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/text-truncate-ellipsis.md - `vault_delete` — Destructive — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/vault-delete.md - `arena_call` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/arena-call.md - `bitwise` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/bitwise.md - `browse` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse.md - `browse_back` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-back.md - `browse_click` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-click.md - `browse_close` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-close.md - `browse_evaluate` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-evaluate.md - `browse_fill` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-fill.md - `browse_navigate` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-navigate.md - `browse_open` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-open.md - `browse_select` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-select.md - `browse_solve_challenge` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-solve-challenge.md - `browse_type` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-type.md - `browse_wait_for` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/browse-wait-for.md - `coach_mission` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/coach-mission.md - `dice_notation_roll` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/dice-notation-roll.md - `gen_avatar_video` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/gen-avatar-video.md - `gen_id_portrait` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/gen-id-portrait.md - `gen_image` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/gen-image.md - `gen_video` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/gen-video.md - `hmac` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/hmac.md - `markdown_to_html` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/markdown-to-html.md - `matrix` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/matrix.md - `regex_tester` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/regex-tester.md - `request_handoff` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/request-handoff.md - `retry_backoff` — Execute — https://policylayer.com/tools/com-wingmanprotocol-agent-gateway/retry-backoff.md - …and 240 more: https://policylayer.com/tools/com-wingmanprotocol-agent-gateway.md ## For agents This record is a snapshot. Live verdicts and the full registry: - Check every server in your MCP config at once: `npx -y policylayer stack` - Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md) - Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events` --- Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=com-wingmanprotocol-agent-gateway · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/com-wingmanprotocol-agent-gateway