Focus.AI Labs / Artifact Field Notes Tools & Utilities
Fast browse / jump off / source-grounded
Tools & Utilities

opencode-dynamic-context-pruning

Context optimization system for extended conversations

Origin
seed-list
Extensibility
standalone
Primary interface
unknown

What this artifact is for

The quick orientation layer before you go deeper upstream.

Profile

Summary

Context optimization system for extended conversations

Seed description: Context optimization system for extended conversations

Category: Tools & Utilities

Best for: Not yet synthesized.

Signals

How to read this page

Workflow tags: none captured

Integrations: none captured

Caveats: No caveats captured yet

Maturity: No maturity signals captured

README excerpt

More of the actual upstream surface so you can skim here before deciding to open the repo.

Upstream skim

Automatically reduces token usage in OpenCode by managing conversation context.

opencode plugin @tarquinen/opencode-dcp@latest --global

This installs the package and adds it to your global OpenCode config.

Preview source: captured upstream text ↗

What shows up in the source

Quick topics

Dynamic Context Pruning PluginInstallationHow It WorksCompressDeduplicationPurge ErrorsConfigurationCommands

Skim bullets

  • range mode compresses contiguous spans of conversation into one or more summaries.
  • message mode (experimental) compresses individual raw messages independently, letting the model manage context much more surgically.
  • Global: ~/.config/opencode/dcp.jsonc (or dcp.json), created automatically on first run
  • Custom config directory: $OPENCODECONFIGDIR/dcp.jsonc (or dcp.json), if OPENCODECONFIGDIR is set
  • Project: .opencode/dcp.jsonc (or dcp.json) in your project's .opencode directory
  • /dcp — Shows available DCP commands
  • /dcp context — Shows a breakdown of your current session's token usage by category (system, user, assistant, tools, etc.) and how much has been saved through pruning.
  • /dcp stats — Shows cumulative pruning statistics across all sessions.

Sources and provenance

Everything on this page should be traceable back to the list seed or an upstream surface.

Provenance

Discovery reason: Listed in the awesome-pi-agent README

Discovered from: seed artifact

Claim confidence: medium

Surface capture kind: repo-readme