Automated README summary
Automatically summarized from the README on 20 Sept 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
An experimental Codex plugin and npm library that wraps Codex's built-in compaction. It prunes tool calls and results with Jev decisions, then re-injects the verbatim kept history as additionalContext. The README warns against using it in real workflows.
How it uses Jev
In PreCompact, the rollout JSONL is replayed and converted to the library's message model. compactMessages() pairs each tool_use with its result, pins recent messages, fits the conversation into maxStateTokens, and asks Jev two noul questions per non-pinned call: should the call stay, should the full result stay verbatim. Decisions and stats are written to $PLUGIN_DATA.
Primitives:noul
Technique worth stealing
Wrap compaction instead of intercepting it: prune only tool calls/results with Jev, keep text verbatim, re-inject after the built-in summary.
Try it
npm install fast-jev-compaction; export TYPESAFE_API_KEY=...; or add the Codex plugin marketplace and trust hooks.
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 material0.77
- Shows a System One patternFound in material0.64
- Describes handling uncertaintynot found in the excerpt0.25
- Reports measured numbersnot found in the excerpt0.05
- Is a runnable toolFound in material0.96
- Top-10 learning shortlist signalnot found in the excerpt0.20
- Is a replica of the modelnot found in the excerpt0.33
- Solves a common problem (0–2)score on a 0–2 scale0.73 / 2
- About JevFound in material0.97
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 20 Sept 2026.