Automated README summary
Automatically summarized from the README on 3 Nov 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
A Docker service with a web UI that classifies new Paperless-ngx inbox documents — document type, correspondent, storage path, issue date and tags — using TypeSafe Jev, or locally via Ollama. Includes a review queue, webhook/polling, and a phone scanner.
How it uses Jev
Each document is sent to Jev as one request containing all questions; Jev chooses from the user's existing Paperless metadata, or from dates the code found in the OCR text. It returns a calibrated confidence per answer: values above the auto threshold are written to Paperless, values above the review threshold go to the review queue, lower ones are ignored.
Primitives:choice
Technique worth stealing
Send all per-document questions in a single Jev request and route each answer by its calibrated confidence into auto-apply, review, or ignore.
Try it
docker run -d --name paperless-jev -p 8090:8000 -v paperless-jev-data:/data ghcr.io/pendraga/paperless-jev:latest
Model judgments from README excerpts
Project signals use repository metadata, an existing summary and the first 700 README characters, not the full project. “Not found” means not detected in that material; it does not establish absence. These are the original judgments with recorded models and dates, not source audits or run results.
Models & evaluation dates
- Inclusion gate
jev-1.14.0- Project signals
jev-1.14.0- Category
jev-1.14.0- Pattern tags
jev-1.14.0
Evidence
These signals use repository metadata, an existing summary and at most 700 cleaned README characters. ≥ 0.60 means reported in that excerpt; ≤ 0.40 means not found there. The middle band is inconclusive. None is a source-code audit or a test result. The runnable-tool question excludes research experiments; the recommendation question asks for a learning shortlist, not a quality verdict.
- Jev is core to the projectnot found in the excerpt0.17
- Shows a System One patternnot found in the excerpt0.11
- Describes handling uncertaintynot found in the excerpt0.15
- Reports measured numbersnot found in the excerpt0.02
- Is a runnable toolFound in material0.98
- Top-10 learning shortlist signalnot found in the excerpt0.18
- Is a replica of the modelnot found in the excerpt0.08
- Solves a common problem (0–2)score on a 0–2 scale1.02 / 2
- About JevFound in material0.98
no-judgment band (0.40–0.60)
no-judgment band (0.40–0.60)
The Jev model and date for each judgment stage are listed above. Card written by DeepSeek V4.1 Flash from the README on 3 Nov 2026.