azdaja
kubet
Bare, open-source RLM layer for existing coding agents
- License
- MIT
- GitHub Stars
- 9
- Source reviewed
- —
Where Jev makes a decision
Jev receives selected source material and explicit questions. It can rank passages, classify records, assess whether evidence supports a claim and judge whether records match for semantic joins. Its returned probabilities let the RLM compare results and decide what to inspect, combine or explore next.
What this project offers
Adds structured choices or scores to the workflow; performance and cost benefits have not been independently verified.
Review scope
Author and repository text is preserved where clear; missing languages are enriched automatically. Checks establish source-level Jev integration, not runtime, safety, or performance validation.
Sources and implementation
- kubet/azdaja/blob/0ac5f0de0c99a5db0d32fa8691dbf08c14663923/README.md
- kubet/azdaja/blob/0ac5f0de0c99a5db0d32fa8691dbf08c14663923/examples/jev_batch_review.py
- kubet/azdaja/blob/0ac5f0de0c99a5db0d32fa8691dbf08c14663923/examples/jev_review_queue.py
- kubet/azdaja/blob/0ac5f0de0c99a5db0d32fa8691dbf08c14663923/src/judge_batch.rs
- kubet/azdaja/blob/0ac5f0de0c99a5db0d32fa8691dbf08c14663923/src/judge.rs
Related projects
Winnow
GhalebDweikat · Context GC
A context filter for Claude Code that hides irrelevant tool-output blocks and keeps the original text available for recall.
fast-jev-compaction
tamaratran · Context GC
Prunes old tool calls and outputs for Claude Code while retaining surviving text verbatim.
jev-skill-gate
ShivamPansuriya · Context GC
Ranks Claude Code skills for the current project to reduce the descriptions loaded by default.
pi-fast-jev-compaction
joelhooks · Context GC
A Pi extension that prunes stale tool history verbatim and falls back to Pi summary compaction when needed.