Automated README summary
Automatically summarized from the README on 18 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
A portable context-compaction layer for coding agents. Instead of asking another LLM to rewrite old context into a lossy summary, it uses Jev to decide which tool calls and results still matter, keeping user and assistant text verbatim across multiple agent runtimes.
How it uses Jev
Jev scores completed tool calls from the transcript state, returning keep probabilities. Tool calls above the keep threshold are retained with full or bounded results; others are removed with their results. The retained normalized context replaces or supplements the host's built-in compaction.
Primitives:score
Technique worth stealing
Ask a calibrated scorer which tool calls still matter instead of asking an LLM to summarize, preserving text verbatim.
Try it
npm install && npm run check, set TYPESAFE_API_KEY, then run save-token-jev compact --format anthropic --input transcript.json
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-1.13.0- Project signals
jev-1.13.0- Category
jev-1.13.0- Pattern tags
jev-1.13.0
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 projectFound in material0.84
- Shows a System One patternnot found in the excerpt0.17
- Describes handling uncertaintynot found in the excerpt0.05
- Reports measured numbersnot found in the excerpt0.04
- Is a runnable toolFound in material0.93
- Top-10 learning shortlist signalunclear0.41
- Is a replica of the modelnot found in the excerpt0.06
- Solves a common problem (0–2)score on a 0–2 scale1.39 / 2
- About JevFound in material0.95
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 18 Oct 2026.