# ai_chat

Interactive chat with AI models

Agent View of the PolicyLayer registry record for `ai_chat`. HTML page: https://policylayer.com/tools/0x-hewm-percepta-mcp/ai-chat

## Facts

- Tool: `ai_chat`
- Server: Percepta MCP Server (`0x-hewm/percepta-mcp`) — https://policylayer.com/tools/0x-hewm-percepta-mcp.md
- Homepage: https://github.com/0x-hewm/percepta-mcp
- Risk category: Execute (High risk)
- Registry record: grade F, identity unverified
- Server rate-limited: no
- Parameters: 0
- Recommended policy verdict: Rate-limited

## Example call (MCP tools/call, JSON-RPC 2.0)

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "ai_chat",
    "arguments": {}
  }
}
```

## Why ai_chat is rated High

Sending messages to AI models constitutes an external operation (API call to OpenAI, Anthropic, Google, or Ollama per server description). The results and side effects depend entirely on the prompt/arguments provided. While read-like in simple cases, it can be used to orchestrate other tools or generate content, placing it in Execute.

From the tool's own definition: "'Interactive chat with AI models' — triggers external AI model API calls whose effects depend on the input arguments"

## Use case

AI agents invoke ai_chat to trigger actions in Percepta MCP Server. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.

## Recommended policy (PolicyLayer)

Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches Percepta MCP Server:

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "ai_chat": {
      "limits": [
        {
          "counter": "ai_chat_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}
```

## Other tools on Percepta MCP Server (21)

- `ai_analyze` — Execute — https://policylayer.com/tools/0x-hewm-percepta-mcp/ai-analyze.md
- `ai_generate` — Execute — https://policylayer.com/tools/0x-hewm-percepta-mcp/ai-generate.md
- `browser_click` — Execute — https://policylayer.com/tools/0x-hewm-percepta-mcp/browser-click.md
- `browser_fill` — Execute — https://policylayer.com/tools/0x-hewm-percepta-mcp/browser-fill.md
- `browser_navigate` — Execute — https://policylayer.com/tools/0x-hewm-percepta-mcp/browser-navigate.md
- `browser_wait_for_element` — Execute — https://policylayer.com/tools/0x-hewm-percepta-mcp/browser-wait-for-element.md
- `execute_test_case` — Execute — https://policylayer.com/tools/0x-hewm-percepta-mcp/execute-test-case.md
- `generate_and_execute_test` — Execute — https://policylayer.com/tools/0x-hewm-percepta-mcp/generate-and-execute-test.md
- `generate_test_case` — Execute — https://policylayer.com/tools/0x-hewm-percepta-mcp/generate-test-case.md
- `monitor_and_analyze_page` — Execute — https://policylayer.com/tools/0x-hewm-percepta-mcp/monitor-and-analyze-page.md
- `start_devtools_monitoring` — Execute — https://policylayer.com/tools/0x-hewm-percepta-mcp/start-devtools-monitoring.md
- `analyze_devtools_anomalies` — Read — https://policylayer.com/tools/0x-hewm-percepta-mcp/analyze-devtools-anomalies.md
- `analyze_image` — Read — https://policylayer.com/tools/0x-hewm-percepta-mcp/analyze-image.md
- `browser_extract_text` — Read — https://policylayer.com/tools/0x-hewm-percepta-mcp/browser-extract-text.md
- `browser_screenshot` — Read — https://policylayer.com/tools/0x-hewm-percepta-mcp/browser-screenshot.md
- `collect_devtools_data` — Read — https://policylayer.com/tools/0x-hewm-percepta-mcp/collect-devtools-data.md
- `compare_images` — Read — https://policylayer.com/tools/0x-hewm-percepta-mcp/compare-images.md
- `extract_structured_data` — Read — https://policylayer.com/tools/0x-hewm-percepta-mcp/extract-structured-data.md
- `extract_text_from_image` — Read — https://policylayer.com/tools/0x-hewm-percepta-mcp/extract-text-from-image.md
- `scrape_website` — Read — https://policylayer.com/tools/0x-hewm-percepta-mcp/scrape-website.md
- `generate_devtools_report` — Write — https://policylayer.com/tools/0x-hewm-percepta-mcp/generate-devtools-report.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=0x-hewm-percepta-mcp · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/0x-hewm-percepta-mcp