Automated README summary
Automatically summarized from the README on 19 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
A DeepSeek Harness plugin that adds four Jev (TypeSafe System One) judgment gates to the agent loop: pre-step, pre-execute, post-execute, and turn-stopping. For Harness users who want calibrated checks on prompts and tool calls.
How it uses Jev
Jev judges four decisions: whether a request is underspecified (pre-step), whether a tool call is destructive, exfiltration, or off-task (pre-execute), whether a successful call silently missed (post-execute), and whether the request is done (turn-stopping). State is the recent transcript or tool call name/arguments/output, capped at maxStateChars. The package owns thresholds and maps probabilities to inject, log/ask/deny, log/block, or nudge/pass; all failures fail open.
Primitives:choicescore
Technique worth stealing
Treat Jev as a calibrated probability source and keep thresholds, decisions, and fail-open behavior in the host package, not the model.
Try it
dsh plugin --profile add dsh-jev-loop, set TYPESAFE_API_KEY, then run dsh --profile jev-core.
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 material0.86
- Shows a System One patternunclear0.59
- Describes handling uncertaintynot found in the excerpt0.05
- Reports measured numbersnot found in the excerpt0.03
- Is a runnable toolFound in material0.89
- Top-10 learning shortlist signalnot found in the excerpt0.24
- Is a replica of the modelnot found in the excerpt0.04
- Solves a common problem (0–2)score on a 0–2 scale1.07 / 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 19 Oct 2026.