Skip to content

[WRONG BRANCH] fix: keep hidden reasoning out of transparent envelopes - #168

Draft
luvs01 wants to merge 1 commit into
mainfrom
codex/fix-hidden-reasoning-exposure-vulnerability
Draft

[WRONG BRANCH] fix: keep hidden reasoning out of transparent envelopes#168
luvs01 wants to merge 1 commit into
mainfrom
codex/fix-hidden-reasoning-exposure-vulnerability

Conversation

@luvs01

@luvs01 luvs01 commented Aug 8, 2026

Copy link
Copy Markdown
Owner

Motivation

  • The previous change encoded suppressed thinking text into a base64 ocxr1: envelope and later decoded it, which exposed hidden chain-of-thought to clients and downstream adapters.
  • The goal is to preserve opaque provider state (signatures, redacted blobs, Kiro ciphertext) without leaking suppressed plaintext reasoning or restoring it into assistant messages.

Description

  • Stop including plaintext txt in reasoning envelopes by removing txt from ReasoningEnvelope and ensuring encodeReasoningEnvelope/decodeReasoningEnvelope only carry sig, red, and krc as opaque payloads (see src/responses/reasoning-envelope.ts).
  • Prevent the bridge from serializing suppressed thinking or raw reasoning text into encrypted_content when hideThinkingSummary is enabled, while keeping the signature and redacted blocks round-tripable (see src/bridge.ts).
  • Stop restoring legacy plaintext envelope txt into assistant thinking content during parsing so legacy/plaintext envelopes no longer become replayed assistant thinking (see src/responses/parser.ts).
  • Preserve internal replay behavior for tool-call replay (replay cache kept) while ensuring hidden text is not emitted in responses; update tests to assert no hidden-text leakage and preserve signature/redaction semantics (tests updated in tests/anthropic-thinking-signature.test.ts and tests/bridge-raw-reasoning-hidden.test.ts).

Testing

  • Ran bun run typecheck and bun run privacy:scan, both succeeded.
  • Ran targeted tests with bun test tests/anthropic-thinking-signature.test.ts tests/bridge-raw-reasoning-hidden.test.ts tests/bridge.test.ts, which exercise the ocxr1 envelope, hidden reasoning, and bridge behavior; the modified tests covering hidden-text leakage and envelope handling passed; the test run reported one environment-specific failure unrelated to these changes (node:zlib.zstdDecompressSync missing in the active Bun runtime).
  • Performed broader bun run test locally to validate integration scope; the changed areas exercised by the suite passed, with an unrelated external-proxy timeout observed in the long-running portion of the full suite.

Codex Task

@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, add credits to your account and enable them for code reviews in your settings.

@coderabbitai

coderabbitai Bot commented Aug 8, 2026

Copy link
Copy Markdown

Warning

Review limit reached

@luvs01, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 10 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details
⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 8d5bb94b-11ef-4e1e-93fe-3fd096678780

📥 Commits

Reviewing files that changed from the base of the PR and between 57140d6 and a9d9aec.

📒 Files selected for processing (5)
  • src/bridge.ts
  • src/responses/parser.ts
  • src/responses/reasoning-envelope.ts
  • tests/anthropic-thinking-signature.test.ts
  • tests/bridge-raw-reasoning-hidden.test.ts

Comment @coderabbitai help to get the list of available commands.

@github-actions

github-actions Bot commented Aug 8, 2026

Copy link
Copy Markdown

Deterministic PR hygiene checks passed.

@github-actions github-actions Bot added the bug Something isn't working label Aug 8, 2026
@github-actions github-actions Bot changed the title fix: keep hidden reasoning out of transparent envelopes [WRONG BRANCH] fix: keep hidden reasoning out of transparent envelopes Aug 8, 2026
@github-actions

github-actions Bot commented Aug 8, 2026

Copy link
Copy Markdown

⏳ DRAFT

  • wrong target branch (main); retarget to dev.

What to do

  • Retarget this PR to dev — all contributions go to dev.

Its title has been prefixed with [WRONG BRANCH].
This pull request was already a draft. Its draft status will be preserved after every issue above is resolved.

@github-actions
github-actions Bot marked this pull request as draft August 8, 2026 11:17
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

aardvark bug Something isn't working codex

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant