Skip to content

Latest commit

 

History

History
43 lines (26 loc) · 2.03 KB

PRODUCT.md

File metadata and controls

43 lines (26 loc) · 2.03 KB

Product

Platform

web

Stack

The product is a TypeScript OpenCode V2 plugin. Its GitHub Pages developer guide uses dependency-free static HTML, CSS, and JavaScript, approved for this surface, deployed with GitHub Actions.

Users

Developers evaluating or installing context compaction for long, tool-heavy OpenCode coding sessions. The developer-guide audience and purpose were confirmed during setup.

Product Purpose

Use a decision model to classify completed tool calls and outputs for retention, rather than asking a generative model to rewrite conversation history. Help site visitors understand this mechanism and install the plugin.

Capabilities and Constraints

  • Experimental OpenCode V2 plugin, version 0.2.0 in the repository.
  • Hosted TypeSafe JEV and an LM Studio-compatible Jev-Style Qwen3.5 2B v1 classifier are supported.
  • The Jev-Style 0.8B v3 dedicated scoring runtime is not directly integrated.
  • Retained text is copied into a deterministic checkpoint; OpenCode's current summary-string hook flattens the message structure.
  • The first and recent messages are pinned; classifier failure or insufficient reduction falls back to native compaction.
  • The public site must require no credentials, send no classifier requests, and explicitly label demonstration data as illustrative.

Evidence on Hand

README.md is the factual source of truth for setup, supported backends, defaults, limitations, troubleshooting, and credits. No benchmark, customer, or measured savings claims are available.

Brand Commitments

Preserve the repository name fast-jev-compaction-opencode and experimental status. No incumbent frontend, logo, or visual identity exists. The user delegated a distinctive technical editorial direction for the site.

Product Principles

  • Explain retention without implying generative summarization.
  • Surface conservative behavior and limitations alongside installation.
  • Keep backend compatibility explicit.
  • Prefer runnable, copyable instructions over unsupported claims.