OrcView

See where your
agent run went wrong.

Inspect tool calls, failures, explicit agent hierarchies, and recorded metrics. Let the AI model you already use explain the evidence.

Try an example ↓Add samples to VS Code ↗

No sign-in. No extra API key. No personal uploads.

One prompt. Real tool output.

6 inspection tools

Two synthetic runs with a nested agent, a failed tool, a retry, and a guardrail event. Choose an example to make a real MCP call. This page shows the evidence; your AI client supplies the conversational answer.

EXAMPLE PROMPTS
Ready when you areEXAMPLE DATA ONLY
Choose a prompt on the left.No model call or API key is needed.

Connect once. Ask naturally.

Add this Streamable HTTP server in an MCP-compatible AI client. It exposes the same public example data as the demo above. Your normal AI account and usage apply.

https://site--orcview--764vq9x779wv.code.run/mcp/orcview/server
Add to VS Code ↗
Configuration for VS Code and compatible clients
{
  "servers": {
    "orcview": {
      "type": "http",
      "url": "https://site--orcview--764vq9x779wv.code.run/mcp/orcview/server"
    }
  }
}

Other clients may use a different configuration format. Choose HTTP / Streamable HTTP and paste the endpoint above. No authorization header is required.

Then ask your model

Bring the tools to your own work.

Existing project-scoped Codex and Claude Code logs, read-only. Content is omitted unless you request it.

uvx --from https://site--orcview--764vq9x779wv.code.run/mcp/install/orcview-0.2.1-py3-none-any.whl orcview-mcp setup --project-root /absolute/path/to/project --cli both --install

Requires Python 3.11+ · uv / uvx · an existing Codex or Claude Code installation.

This command registers OrcView with your installed Codex CLI and Claude Code. Use --cli codex or --cli claude for one. It preserves existing same-name registrations. Restart your client after setup. Other IDEs can use the stdio command shown below.

Stdio command for another IDE
uvx --from https://site--orcview--764vq9x779wv.code.run/mcp/install/orcview-0.2.1-py3-none-any.whl orcview-mcp --project-root /absolute/path/to/project --cli both
Download the versioned install package ↗
What “local” means for privacy

These MCP tools do not save request or response content, enable telemetry, or call a model provider. They return selected evidence to your AI host. That host may retain it or send it to its model provider under its own policy.

OrcView reads logs already retained by the CLI; it does not make a second recording store. Its adapters cannot expose internals that the CLI did not log. Billed cost and provider latency remain unknown for imported CLI runs.