feat(vision): add optional OmniParser HTTP provider (#1053)#1107
Open
shaun0927 wants to merge 2 commits into
Open
feat(vision): add optional OmniParser HTTP provider (#1053)#1107shaun0927 wants to merge 2 commits into
shaun0927 wants to merge 2 commits into
Conversation
Constraint: OmniParser must remain an external, explicitly configured HTTP service so core OpenChrome stays dependency-light. Rejected: Add Python/Torch/Docker or make OmniParser the default provider | violates the adapter-only issue boundary and would risk harness stability. Confidence: high Scope-risk: moderate Directive: Keep external visual providers timeout-bounded, warning-backed, and safe to fall back to DOM snapshots. Tested: npm test -- --runInBand tests/vision/omniparser-http-provider.test.ts; npm test -- --runInBand tests/vision/perception-snapshot.test.ts tests/vision/omniparser-http-provider.test.ts; npm run build; npm run lint:tier; npm run lint:changed; git diff --check Not-tested: Full npm test not rerun because the preceding PR branch already exposed unrelated tests/tools/tabs.test.ts failures; npm run lint still has unrelated baseline errors in src/session-manager.ts and src/tools/connect.ts; live OpenChrome MCP mock-server verification not run in this worktree.
|
Warning You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again! |
Qodo reviews are paused for this user.Troubleshooting steps vary by plan Learn more → On a Teams plan? Using GitHub Enterprise Server, GitLab Self-Managed, or Bitbucket Data Center? |
Resolve the stacked vision_find conflict so the optional OmniParser HTTP snapshot provider coexists with occlusion, iframe, and tiled DOM capture options inherited from the perception snapshot base. Constraint: PR #1107 targets the perception-snapshot branch after that branch learned richer DOM capture options. Rejected: Dropping tiled/iframe options or the OmniParser provider | both are opt-in surfaces and do not need to conflict. Confidence: medium Scope-risk: moderate Directive: Keep OmniParser HTTP opt-in and preserve legacy DOM output as the default vision_find behavior. Tested: npx jest tests/vision/omniparser-http-provider.test.ts tests/vision/vision-find.test.ts tests/vision/perception-snapshot.test.ts --runInBand --forceExit; npm ci; npm run build; git diff --check for touched file. Not-tested: Full GitHub Actions matrix. Co-authored-by: OmX <omx@oh-my-codex.dev>
Open
27 tasks
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
1 participant
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.
Summary
Implements #1053 as a stacked follow-up on #1093 / #1052:
OmniParserHttpProvider, an optional HTTP adapter for an already-running OmniParser-compatible serviceOPENCHROME_VISION_PROVIDER=omniparser-httpand snapshot output is requestedOPENCHROME_OMNIPARSER_URL,OPENCHROME_OMNIPARSER_TIMEOUT_MS, andOPENCHROME_OMNIPARSER_MAX_ELEMENTSparsed_content_listratio/pixel bboxes intoPerceptionSnapshotelementsStack / duplicate check
feat/1052-perception-snapshotbecause this depends on feat(vision): add provider-neutral perception snapshots (#1052) #1093's provider-neutral snapshot contract.vision_find; this PR adds only an optional external provider adapter.package.json.Verification
Passed:
npm test -- --runInBand tests/vision/omniparser-http-provider.test.tsnpm test -- --runInBand tests/vision/perception-snapshot.test.ts tests/vision/omniparser-http-provider.test.tsnpm run buildnpm run lint:tiernpm run lint:changedgit diff --checkBaseline notes:
npm run lintstill fails on unrelated existing errors:src/session-manager.ts:959:7no-useless-catchsrc/tools/connect.ts:43:69@typescript-eslint/ban-typesnpm testwas not rerun here because the preceding feat(vision): add provider-neutral perception snapshots (#1052) #1093 worktree already exposed unrelated existing failures intests/tools/tabs.test.ts.Merge validation path via OpenChrome
After #1093 lands, validate this PR with an OmniParser-compatible mock service:
OPENCHROME_VISION_PROVIDER=omniparser-httpOPENCHROME_OMNIPARSER_URL=http://127.0.0.1:<mock-port>/parse/vision_findwith{ "format": "snapshot", "includeImage": false }.provider: "omniparser-http",source: "omniparser-http"elements, bounded labels, and normalizedbboxRatio.omniparser-httpdoes not appear in default snapshot output.Closes #1053.