# Opencode-DCP/opencode-dynamic-context-pruning

Dynamic context pruning plugin for OpenCode - intelligently manages conversation context to optimize token usage

Repository: https://github.com/Opencode-DCP/opencode-dynamic-context-pruning
Canonical: https://ross.abutalabs.com/products/opencode-dynamic-context-pruning
Language: TypeScript
License: AGPL-3.0
License Family: copyleft
Last push: 2026-08-16T04:10:39+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 98, release rhythm 86, longevity 20
- inputs: {"age_days": 287, "days_push": 17, "days_rel": 17, "gap_med": 1, "n_releases_24m": 68}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4085, forks 228 (observed 2026-08-28T04:08:34.442583+00:00)

## What it is
A plugin for OpenCode that dynamically prunes and compresses conversation context to reduce token usage when working with LLMs. It replaces stale or redundant content with placeholders and summaries without modifying session history.

## Use cases
- reduce token usage in long OpenCode coding sessions
- compress stale conversation history into summaries
- deduplicate repeated tool call outputs
- prune errored tool call inputs automatically
- avoid hitting context window limits during agent tasks

## When to choose
- you use OpenCode and want automatic context management
- your sessions grow too large and hit token limits
- you want smarter, model-driven compression instead of static compaction

## When to avoid
- you are starting fresh - the maintainers recommend the Sleev CLI instead
- you need a harness-agnostic proxy for Claude Code or Codex
- you need verbatim preservation of all large prompts

## Facets
- artifact type: plugin
- maturity: maintenance
- function: llm-inference, middleware, plugin-system
- domain: large-language-models, developer-tools, artificial-intelligence
- platform: cli
- tags: context-management, token-optimization, opencode, context-pruning, prompt-caching, nodejs

## Member repositories
- Opencode-DCP/opencode-dynamic-context-pruning (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:34.442583+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:23:23.896456+00:00, confidence not recorded.
  - readme: https://github.com/Opencode-DCP/opencode-dynamic-context-pruning (fetched 2026-08-28T04:08:34.442583+00:00, sha d1682f85d5ed)
- Data as of 2026-08-30T08:39:29.467469+00:00.