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
Doc is a Python toolkit that plugs TypeSafe's Jev model into Claude Code, Codex, Grok and Hermes as hooks, adding small typed checks to ordinary agent work: guarding published comments, screening final replies, routing skills, and picking a cheaper model.
How it uses Jev
Jev answers typed questions on the turn's state: publish_guard scores comment drafts for local-only state, effort estimates, process narration and billing; reply_check asks whether a reply gives an effort estimate or requests a secret; skill_router ranks skills against the turn. Scores drive deny/ask, block-once, or a suggested-skill line.
Primitives:choicescorenoul
Technique worth stealing
Log every Jev call and decision with its scores to a JSONL file, never the judged text, so thresholds can be tuned on real traffic.
Try it
Register the hooks in ~/.claude/settings.json or ~/.codex/hooks.json, then link the CLI with ln -s ~/doc/bin/doc /usr/local/bin/doc.
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 projectFound in material0.91
- Shows a System One patternFound in material0.81
- Describes handling uncertaintynot found in the excerpt0.34
- Reports measured numbersnot found in the excerpt0.02
- Is a runnable toolFound in material0.95
- Top-10 learning shortlist signalunclear0.52
- Is a replica of the modelnot found in the excerpt0.03
- Solves a common problem (0–2)score on a 0–2 scale1.84 / 2
- About JevFound in material0.99
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.