fix(ml): deliver verified forecasts and portfolio opportunity analysis - #67
Merged
Merged
Conversation
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
There was a problem hiding this comment.
Your trial has ended. Reactivate Greptile to resume code reviews.
added 6 commits
October 2, 2026 13:14
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The ML screens previously mixed missing results, stale observations and descriptive signals with apparent predictions. This release returns explicit availability, dates and provenance, preserves the selected portfolio source, and publishes only the volatility artifacts accepted by the retrospective validation protocol.
Market Opportunities now explores historical candidates independently of indeterminate sector gaps. It retains unknown fund composition and missing valuations, explains exclusions, and offers manual scenarios only when the selected snapshot can be valued. The UI uses styled tables, summary cards and expandable evidence. No automatic buy/sell decision, forward-return claim or allocation integration is added.
The release includes verified public daily datasets, 62 JSON risk artifacts and evaluation reports. Exact-byte Git attributes preserve their receipts and code fingerprints across Windows/Linux checkout. No user portfolio, cash record, key, session, private screenshot or .env file is included. Unrelated local changes remain outside these commits.
Validation: 160 frontend tests and 18 focused history/analytics tests pass locally; prior backend and authenticated preview results are documented in docs/audit. The PR CI and final Linux/production checks must pass before publication. The robot2 delivery script preserves the existing production configuration, creates a private host-only backup, checks write paths on a local copy and retains the previous container for rollback.
Robot2 final candidate validation: 156 Linux tests passed (14 skipped), 18 legitimate-write/authentication controls passed on a private host-only copy, and the authenticated Cycle Analysis, AI Dashboard, Advanced Analytics, Admin and Market Opportunities pages passed at desktop/mobile widths. The selected Saxo CSV and saved cash remain coherent. Public observations are refreshed independently of immutable evaluation datasets; automatic model training remains disabled. Both Dockerfiles preserve this policy for future rebuilds.
Existing GitHub Dependabot findings on npm development tools remain outside this release; the project's npm toolchain and node_modules directory are absent from the production candidate. CI passed on the exact PR head: 3504 backend tests passed (25 skipped), 160 frontend tests passed across 16 suites, lint and Python dependency/security audits passed.