mirror of
https://github.com/openai/codex.git
synced 2026-10-30 19:18:49 +00:00
jif
9def09afa7
## What changed - Match final serving assessments to their response and output item, then retain them with tool invocations and code cells across waits. - Use matching assessments instead of asynchronous scoring for adaptive review: low-risk actions can skip synchronous review, while elevated-risk actions require it. - Revalidate captured authorization and policy before approval. Require fresh review for stale or incomplete authorization, preserve explicit review requirements, and discard assessments when hooks rewrite tool input. - Keep asynchronous scoring as the fallback when an assessment is missing or does not match. ## Testing Add integration coverage for direct and nested tool calls, hook rewrites, explicit escalation, interrupted responses, catalog-driven policy, incomplete authorization, and assessment retention across code-cell yields and waits. GitOrigin-RevId: 4346a227204137f018c03c34b302d6f1e9592c96
37 lines
1.4 KiB
Rust
37 lines
1.4 KiB
Rust
//! Decodes action-bound Guardian metadata without granting permission to execute.
|
|
//! Missing or malformed assessments remain unavailable; consumers must validate the
|
|
//! response, completed action, and current authorization before using a low-risk result.
|
|
|
|
use serde_json::Value;
|
|
|
|
/// One final serving assessment emitted immediately before the matching output item is done.
|
|
/// It is carried outside model-visible response items and is not subsequently revised.
|
|
///
|
|
/// Consumers must validate the response, completed action, and current authorization
|
|
/// before using a low-risk result. This assessment does not grant permission to execute.
|
|
#[derive(Clone, Debug, PartialEq, Eq)]
|
|
pub struct GuardianAssessment {
|
|
pub response_id: String,
|
|
pub item_id: String,
|
|
/// False only permits considering the action for the Guardian fast path.
|
|
pub requires_review: bool,
|
|
}
|
|
|
|
pub(crate) fn parse_guardian_assessment(
|
|
response_id: Option<&str>,
|
|
metadata: Option<&Value>,
|
|
) -> Option<GuardianAssessment> {
|
|
let response_id = response_id?;
|
|
let guardian = metadata?.get("guardian")?;
|
|
let item_id = guardian.get("item_id")?.as_str()?;
|
|
let requires_review = guardian.get("requires_review")?.as_bool()?;
|
|
if response_id.is_empty() || item_id.is_empty() {
|
|
return None;
|
|
}
|
|
Some(GuardianAssessment {
|
|
response_id: response_id.to_owned(),
|
|
item_id: item_id.to_owned(),
|
|
requires_review,
|
|
})
|
|
}
|