Automated README summary
Automatically summarized from the README on 6 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
TDM turns subjective judgment calls — choices, yes/no judgments, ordered scores — into a typed, auditable API. Callers submit a typed question with decision context and receive a typed answer with a winning label, probability distribution, and confidence. Shipped as an opencode plugin.
How it uses Jev
Jev is the first provider behind TDM's pluggable provider interface. The opencode plugin exposes a tdm_judge tool for typed judgments mid-session (routing, classification, noul yes/no with P(yes), ordered scoring), and a risk gate that initiates a TDM risk judgment (a score plus noul questions over {tool, args}), enforced at permission.evaluate by mutating effect to "ask" when the verdict is dangerous.
Primitives:choicenoulscore
Technique worth stealing
Typed questions with full decision context return typed answers (winning label, probability distribution, confidence) that feed code logic and stay auditable.
Try it
Add "@typedecision/opencode-tdm@latest" to the plugins array in opencode.jsonc and set TYPESAFE_API_KEY.
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 projectnot found in the excerpt0.16
- Shows a System One patternnot found in the excerpt0.38
- Describes handling uncertaintynot found in the excerpt0.07
- Reports measured numbersnot found in the excerpt0.05
- Is a runnable toolFound in material0.98
- Top-10 learning shortlist signalnot found in the excerpt0.20
- Is a replica of the modelnot found in the excerpt0.20
- Solves a common problem (0–2)score on a 0–2 scale1.52 / 2
- About JevFound in material0.87
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 6 Oct 2026.