Repository navigation
feat(mcp): preserve bounded inline image results - #110
Merged
Merged
Conversation
Let hosted desktop tools return the actual screenshot to native MCP harnesses while retaining the text summary. Validate inline formats and bounds and exclude failed results; metadata remains a presentation hint, not authority.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Successful host tools can now return bounded inline images through
metadata['sharedos/content']. The MCP adapter keeps its existing text summary and appends validated JPEG, PNG, or WebP image blocks, allowing a native MCP harness to see an actual desktop screenshot.Images are limited to four blocks and 4 MiB of encoded data in total. Invalid payloads and failed tool results never produce image blocks; no URL fetching, exports, or authority rules change.
Validation: 1,140 tests, typecheck, formatting, API docs check, release checks, and conformance checks passed. A real native Codex harness observed and operated two independent Linux desktops through this adapter.
This PR is stacked on
feat/reasoning-effort, the existing revision already pinned by the downstream runtime. Only the image-result implementation, tests, and documentation are new here.