Scope MCP app resource reads to their originating call (#39187)

## Why

App widgets need to read resources using the app and account context of the
tool call that produced them, including after a thread is restored.

## What changed

- Add an optional `originCallId` to `mcpServer/resource/read` requests and
  return it on successful scoped reads.
- Track bounded provenance for successful app tool calls in thread history and
  use the current tool binding, account link, and app policy when reading the
  associated widget resource.
- Reject missing, mismatched, failed, or ambiguous origins, while keeping
  app-only tools available for widget reads without exposing them to the model.

## Testing

Add app-server coverage for scoped widget reads across both history modes,
ephemeral and persistent threads, server restarts, app-only visibility, and
invalid origin cases.

GitOrigin-RevId: 11eaefc066b2a3b639599e4cbdef680233d150a2
This commit is contained in:
jif
2026-08-18 11:51:21 +00:00
committed by copyberry
parent 711a5f8b3a
commit 880f1135ea
26 changed files with 779 additions and 22 deletions
+17
View File
@@ -747,6 +747,23 @@ impl CodexThread {
Ok(serde_json::to_value(result)?)
}
/// Reads an app resource using the current authority of its originating tool call.
pub async fn read_mcp_resource_for_call(
&self,
call_id: &str,
uri: &str,
) -> anyhow::Result<serde_json::Value> {
self.session.refresh_mcp_if_dirty().await;
let result = self
.session
.services
.mcp_runtime
.read_resource_for_call(self.session.thread_id, call_id, uri)
.await?;
Ok(serde_json::to_value(result)?)
}
pub async fn call_mcp_tool(
&self,
server: &str,