Skip to content

Classify tool calls in the Guardian V2 extension - #38409

Merged
copyberry[bot] merged 1 commit into
mainfrom
copyberry/codex-internal-to-codex-oss/6b69fcc0443086544086849c38bcdd90560a291f
Aug 13, 2026
Merged

Classify tool calls in the Guardian V2 extension#38409
copyberry[bot] merged 1 commit into
mainfrom
copyberry/codex-internal-to-codex-oss/6b69fcc0443086544086849c38bcdd90560a291f

Conversation

@copyberry

@copyberry copyberry Bot commented Aug 13, 2026

Copy link
Copy Markdown

Classify tool calls in the Guardian V2 extension

What changed

  • Register a tool lifecycle contributor when Guardian V2 is installed.
  • On tool start, asynchronously ask the existing Luna sampler for a low-effort
    action_risk score constrained to the range from 0.0 to 1.0.
  • Bound the tool metadata supplied to the classifier and treat it as untrusted
    input.
  • Emit an extension warning if classification sampling fails.

Testing

Add an integration test that verifies the contributor reuses the configured
Luna pool and sends the expected model, attribution metadata, reasoning effort,
strict output schema, and tool-call context.

## What changed

- Register a tool lifecycle contributor when Guardian V2 is installed.
- On tool start, asynchronously ask the existing Luna sampler for a low-effort
  `action_risk` score constrained to the range from `0.0` to `1.0`.
- Bound the tool metadata supplied to the classifier and treat it as untrusted
  input.
- Emit an extension warning if classification sampling fails.

## Testing

Add an integration test that verifies the contributor reuses the configured
Luna pool and sends the expected model, attribution metadata, reasoning effort,
strict output schema, and tool-call context.

GitOrigin-RevId: 6b69fcc0443086544086849c38bcdd90560a291f
@copyberry
copyberry Bot force-pushed the copyberry/codex-internal-to-codex-oss/6b69fcc0443086544086849c38bcdd90560a291f branch from 301179c to d167a36 Compare August 13, 2026 17:32
@copyberry
copyberry Bot merged commit d167a36 into main Aug 13, 2026
28 of 32 checks passed
@copyberry
copyberry Bot deleted the copyberry/codex-internal-to-codex-oss/6b69fcc0443086544086849c38bcdd90560a291f branch August 13, 2026 17:33
@github-actions github-actions Bot locked and limited conversation to collaborators Aug 13, 2026
@X-000-000
X-000-000 deployed to issue-triage August 13, 2026 18:06 — with GitHub Actions Active
@X-000-000
X-000-000 deployed to issue-triage August 13, 2026 18:06 — with GitHub Actions Active
@X-000-000
X-000-000 deployed to issue-triage August 13, 2026 18:06 — with GitHub Actions Active
@dylbarne
dylbarne deployed to issue-triage August 13, 2026 18:08 — with GitHub Actions Active
@dylbarne
dylbarne deployed to issue-triage August 13, 2026 18:08 — with GitHub Actions Active
@dylbarne
dylbarne deployed to issue-triage August 13, 2026 18:08 — with GitHub Actions Active
@dylbarne
dylbarne deployed to issue-triage August 13, 2026 18:09 — with GitHub Actions Active
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants