Skip to content
#

agent-middleware

Here are 2 public repositories matching this topic...

Layered token-optimization pipeline for DeepSeek Harness: output ladder, MCP lazy loading,compaction driver, cache-hit reporting. Built on real DSH plugin APIs; ~40-60% input saved in long sessions.

  • Updated Oct 4, 2026
  • JavaScript

Behavior-management plugin for DeepSeek Harness: tool-call discipline prompt section, failure-triggered parallelism convergence (pool drops to 1, auto-restores), consecutive-failure user intervention. Complements dsh-token-optimizer; works standalone.

  • Updated Oct 2, 2026
  • JavaScript

Add this topic to your repo

To associate your repository with the agent-middleware topic, visit your repo's landing page and select "manage topics."

Learn more