A live key does not make old scores live
The fixed source uses a 30-day cache without a mock/live namespace or invalidation on key changes. Old mock classifications can therefore survive enabling a key.
Inspect the supporting source ↗REPOSITORY ANALYSIS
JevFilterForX
A browser extension that scores and filters X posts in real time with Jev, folding low-signal content while keeping it expandable. Without an API key, it defaults to local mock scoring.
E2 · source reviewedPublic material or source evidence was reviewed. A fixed-version source review requires checking a pinned revision; this is not a reproduction, performance endorsement, or safety guarantee.
- Stars
- 1
- Forks
- 0
- License
- —
- Last commit
- 2026-09-20
What Jev does here
Jev classifies posts with Choice, rates signal, actionability and originality with Score, and assigns Noul labels; local thresholds and noise rules decide whether to fold each post.
Adds scores, labels and adjustable filtering thresholds to the X feed, with reversible folding of posts and their attached media.
Fixed-commit source review only; no runtime, accuracy, latency or cost validation. Mock and live results share a 30-day cache that is not invalidated when a key is added. Interest/value preferences do not affect filtering; missing API scores become zero. Online mode sends post text, author and ID to TypeSafe; the key is stored locally. Public source has no identified license.
What task does this project fit?
Understand an X-feed filter before trusting its scores or enabling a live key.
Material analysed：
From input to action
- Posts are looked up in a shared tweet-ID cache before classification.
- Without a key a text-hash mock scores them; with a key Jev answers value, score and topic/noise questions.
- Local policy uses score and noise settings to fold posts reversibly.
Implementation details that matter
Not every visible setting affects filtering
The inspected policy consumes focus score and selected noise rules. Interest/value preference controls are not evidence that those preferences personalize the actual visibility decision.
Inspect the supporting source ↗When to choose it
Useful for studying reversible ranking and folding. For a meaningful live evaluation, separate or clear mock-derived cache entries and inspect which preferences policy actually reads.
How to inspect it yourself
Trace createClassifier in classifier.ts and the cache lookup in the background pipeline, then compare policy.ts with settings. Expect no key to select MockClassifier and a cache hit to bypass a new classifier call.
Limits of this review
- No extension run or accuracy measurement was performed.
- Online mode transmits post data to TypeSafe; no license was identified in the reviewed record.