COMMUNITY OBSERVEDMetric claims: AUTHOR REPORTEDSOURCE AUDITED🤖 Jev Agent

dsh-jev-prune

Context compaction for DeepSeek Harness, judged by Jev.

Overview & result

yangyu666 shows dsh-jev-prune, built on Jev: Context compaction for DeepSeek Harness, judged by Jev.

From the original post

dsh-jev-prune by @yangyu666: Context compaction for DeepSeek Harness, judged by Jev.

— yangyu666

How Jev fits in the loop

  1. When a long tool result or old history needs reclaiming, the plugin asks Jev yes/no questions about whether each piece is still useful for the current goal.
  2. The state carries the most recent user instruction, so usefulness is judged against the task.
  3. Probabilities are used as a relative ranking within the session, never as a fixed threshold.
  4. Kept text stays verbatim; compacted regions are replaced by a deterministic receipt written by code, not by a model.

How to reproduce

  1. Review the source documentation and repository at https://github.com/yangyu666/dsh-jev-prune.
  2. Configure environment with TypeSafe Jev / System One API keys.
  3. Run test prompts against the decision endpoints.

Why this build matters

Shows Jev in practical use for browser and autonomous agents.

Limitations

  • Measured results come from one 37-step file-reading task, compared across three arms.
  • Write-type calls and the recent window are protected by hard rules, not by probabilities.
  • The author notes a small judge model's output distribution is narrow, so only relative ordering within a session is reliable.

Patterns