fix(openai-chat): bound Moonshot schema inlining bytes by luvs01 · Pull Request #355 · luvs01/opencodex

luvs01 · 2026-08-28T04:10:27Z

Motivation

Prevent a request-amplification DoS where inlining a large $defs properties map into many sibling-bearing $ref nodes could multiply serialized output and exhaust process memory.
Preserve the chosen Moonshot $ref-with-siblings semantics (conjunction for required/properties, node-narrowing otherwise) while making inlining resource-bounded and safe.

Description

Add a serialized-byte budget MOONSHOT_MAX_INLINED_SCHEMA_BYTES (1 MiB) and an iterative estimator serializedJsonBytesUpTo to bound how many bytes of a referenced schema may be inlined. (src/adapters/openai-chat.ts)
Track per-normalization state with remainingInlineBytes and inlineSizeCache so large referenced targets are measured once and charged before being copied, and reuse cached sizes for repeated references. (src/adapters/openai-chat.ts)
When the inlining budget would be exceeded, preserve a Moonshot-compatible bare $ref instead of duplicating the large map, and keep the existing depth/node/expansion guards. (src/adapters/openai-chat.ts)
Add a focused regression test that constructs a 10k-entry boolean properties definition referenced many times and asserts that emitted output stays bounded and that some references remain as bare $ref. (tests/moonshot-tool-schema.test.ts)
Update the adapter registry decision note to document the new serialized-byte budget and its rationale. (structure/10_adapter-registry.md)

Testing

Ran the focused Moonshot suite with bun test tests/moonshot-tool-schema.test.ts, which passed (17 tests, 0 failures).
Ran typecheck with bun run typecheck, which completed without type errors.
Ran the privacy check with bun run privacy:scan, which passed.
Attempting the full repository bun run test was not used for acceptance here because unrelated, pre-existing management/auth tests timed out/failed in this environment; the change was validated by the focused tests above.

Codex Task

github-actions · 2026-08-28T04:10:39Z

✅ Deterministic PR hygiene checks passed.

github-actions · 2026-08-28T04:10:48Z

✅ READY

all PR quality gates passed.

Hygiene

✅ Deterministic PR hygiene checks passed.

coderabbitai · 2026-08-28T04:10:49Z

Important

🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

Configuration used: Repository: luvs01/opencodex/.coderabbit.yaml

Review profile: ASSERTIVE

Plan: Advanced

Run ID: d20ed147-d9b2-4bec-9ff4-bd1c35213f3e

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

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

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: ab7f684bfc

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-08-28T04:13:07Z

  const normalized = normalizeMoonshotSchemaNode(rooted, rooted, {
    activeRefs: new Set<string>(),
+    inlineSizeCache: new WeakMap<Record<string, unknown>, number>(),
    remainingExpansions: MOONSHOT_MAX_REF_EXPANSIONS,
+    remainingInlineBytes: MOONSHOT_MAX_INLINED_SCHEMA_BYTES,


Share the byte allowance across the whole tool catalog

The allowance is initialized inside normalizeMoonshotToolParameters, so toolsToChatFormat grants a fresh 1 MiB amplification budget to every tool. Because the inbound tool arrays have no aggregate count or output-size limit, a catalog containing many individually small schemas can still exhaust memory while constructing and serializing the final request; for example, I reproduced a 128-tool, 3.36 MB catalog expanding to about 82 MB. Allocate the budget once per request/catalog and pass it through each normalization, or enforce an aggregate serialized-output cap.

Useful? React with 👍 / 👎.

devin-ai-integration

Note

This report is out of date. Scroll down for Devin Review's latest report on this PR.

✅ Devin Review: No Issues Found

Devin Review analyzed this PR and found no bugs or issues to report.

…ication Resolve conflicts from the openai-chat module split: port the serialized-byte inline budget into src/adapters/openai-chat/tool-schema.ts, move the adapter-registry note to structure/decisions/ADR-0093, and keep the moved Moonshot regression tests at tests/providers/moonshot-tool-schema.test.ts. Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

…ication

…test.ts) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

Decision records are historical; the byte-budget reasoning moves out of ADR-0093 into ADR-0355 and the chat-compat contract now states the inline-byte bound alongside depth, node, and expansion bounds. Co-Authored-By: Epinephrine <luvs01@hanmail.net>

devin-ai-integration · 2026-09-21T14:31:56Z

Fixed in 3b8cb5e: ADR-0093 restored verbatim; the byte-budget reasoning now lives in a new record, structure/decisions/ADR-0355-chat-structured-output-compatibility.md, and the current contract is stated in structure/providers/chat-compat.md.

devin-ai-integration · 2026-09-21T14:31:58Z

Fixed in 3b8cb5e: the "Chat structured-output compatibility" contract now states the walk is depth-, node-, expansion-, and inline-byte-bounded (shared 1 MiB allowance, bare-$ref fallback on overflow) and links the new ADR-0355.

devin-ai-integration · 2026-09-22T01:37:29Z

Addressed in 3b8cb5e — ADR-0093 is restored byte-for-byte to its original text; the byte-budget rationale now lives in the new ADR-0355 (linked from the owning contract).

devin-ai-integration · 2026-09-22T01:37:32Z

Addressed in 3b8cb5e — the contract now states the walk is depth-, node-, expansion-, and inline-byte-bounded (each inlined reference charged its serialized size against a shared 1 MiB allowance, over-budget keeps the bare-$ref fallback).

…t tools A per-tool 1 MiB allowance let a large catalog multiply the cap by its tool count, reopening the request amplification the budget exists to bound. Hoist the allowance to one MoonshotInlineByteBudget per toolsToChatFormat call so every tool spends from the same pool. Adds a regression test: two tools carrying ~0.6 MB inline targets each now share one allowance, so only the first of four sibling refs inlines and the rest keep the bare-$ref fallback. Co-Authored-By: Epinephrine <luvs01@hanmail.net>

devin-ai-integration · 2026-09-22T08:20:38Z

Fixed in 292e7e1 — the 1 MiB allowance is now one MoonshotInlineByteBudget per toolsToChatFormat call shared by every tool in the request, so a large catalog can no longer multiply the cap by its tool count. Regression test added: two tools each carrying ~0.6 MB inline targets now share the pool, leaving three of four sibling refs as bare $ref.

fix(openai-chat): bound Moonshot schema inlining bytes

ab7f684

luvs01 added aardvark codex labels Aug 28, 2026 — with ChatGPT Codex Connector

github-actions Bot added the bug Something isn't working label Aug 28, 2026

github-actions Bot changed the title ~~fix(openai-chat): bound Moonshot schema inlining bytes~~ [WRONG BRANCH] fix(openai-chat): bound Moonshot schema inlining bytes Aug 28, 2026

github-actions Bot marked this pull request as draft August 28, 2026 04:10

chatgpt-codex-connector Bot reviewed Aug 28, 2026

View reviewed changes

luvs01 changed the title ~~[WRONG BRANCH] fix(openai-chat): bound Moonshot schema inlining bytes~~ fix(openai-chat): bound Moonshot schema inlining bytes Sep 3, 2026

luvs01 changed the base branch from main to dev September 3, 2026 06:23

github-actions Bot marked this pull request as ready for review September 3, 2026 06:33

ci: retrigger checks (empty commit; dev merge conflicts)

060cc36

devin-ai-integration Bot reviewed Sep 21, 2026

View reviewed changes

This comment was marked as resolved.

Sign in to view

luvs01 and others added 3 commits September 21, 2026 08:36

Merge branch 'dev' into codex/propose-fix-for-moonshot-request-amplif…

2f6ab89

…ication

ci: retrigger macos 1/2 (wedged-spawn silent-stall in client-connect.…

7c8e39b

…test.ts) Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

This comment was marked as resolved.

Sign in to view

Navigation Menu

fix(openai-chat): bound Moonshot schema inlining bytes - #355

fix(openai-chat): bound Moonshot schema inlining bytes#355
luvs01 wants to merge 7 commits into
devfrom
codex/propose-fix-for-moonshot-request-amplification

luvs01 commented Aug 28, 2026 •

edited by devin-ai-integration Bot

Loading

Uh oh!

github-actions Bot commented Aug 28, 2026

Uh oh!

github-actions Bot commented Aug 28, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot commented Aug 28, 2026 •

edited

Loading

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

chatgpt-codex-connector Bot Aug 28, 2026

Uh oh!

devin-ai-integration Bot left a comment •

edited

Loading

Uh oh!

This comment was marked as resolved.

Uh oh!

devin-ai-integration Bot commented Sep 21, 2026

Uh oh!

devin-ai-integration Bot commented Sep 21, 2026

Uh oh!

This comment was marked as resolved.

Uh oh!

devin-ai-integration Bot commented Sep 22, 2026

Uh oh!

devin-ai-integration Bot commented Sep 22, 2026

Uh oh!

devin-ai-integration Bot commented Sep 22, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

luvs01 commented Aug 28, 2026 • edited by devin-ai-integration Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Motivation

Description

Testing

Uh oh!

github-actions Bot commented Aug 28, 2026

Uh oh!

github-actions Bot commented Aug 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

✅ READY

Hygiene

Uh oh!

coderabbitai Bot commented Aug 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Aug 28, 2026

Choose a reason for hiding this comment

Uh oh!

devin-ai-integration Bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

✅ Devin Review: No Issues Found

Uh oh!

This comment was marked as resolved.

Uh oh!

devin-ai-integration Bot commented Sep 21, 2026

Uh oh!

devin-ai-integration Bot commented Sep 21, 2026

Uh oh!

This comment was marked as resolved.

Uh oh!

devin-ai-integration Bot commented Sep 22, 2026

Uh oh!

devin-ai-integration Bot commented Sep 22, 2026

Uh oh!

devin-ai-integration Bot commented Sep 22, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

luvs01 commented Aug 28, 2026 •

edited by devin-ai-integration Bot

Loading

github-actions Bot commented Aug 28, 2026 •

edited

Loading

coderabbitai Bot commented Aug 28, 2026 •

edited

Loading

devin-ai-integration Bot left a comment •

edited

Loading