Automated README summary
Automatically summarized from the README on 4 Nov 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
A TypeScript plugin for coding agents that safeguards against dangerous shell commands, improves token efficiency, and delivers more reliable task solutions. It supports eight agents, with OpenCode as the reference implementation where all features work.
How it uses Jev
Jev (or another System One model at /systemone) acts as the classifier. It scores skill fit (noul per skill), qualifies prompt complexity (low/medium/high), judges tool-group relevance, picks a tool via a choice decision in reflex_run, and checks edits and shell commands. Verdicts drive maxTokens, tool exposure, and history compaction.
Primitives:choicescorenoul
Technique worth stealing
Batch classifier questions per request and cache verdicts per session so per-request cost stays flat as sessions grow.
Try it
Configure agent-reflex.jsonc with a System One model at /systemone; OpenCode is the reference implementation.
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 excerpt0.06
- Shows a System One patternnot found in the excerpt0.07
- Describes handling uncertaintynot found in the excerpt0.04
- Reports measured numbersnot found in the excerpt0.02
- Is a runnable toolFound in material0.95
- Top-10 learning shortlist signalnot found in the excerpt0.10
- Is a replica of the modelnot found in the excerpt0.05
- Solves a common problem (0–2)score on a 0–2 scale1.65 / 2
- About JevFound in material0.91
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 4 Nov 2026.