A Claude Code plugin that shows what's happening - context usage, active tools, running agents, and todo progress
SpineCodex
SpineCodex: Let your Codex work, evolve, and scale on a SpineTree — up to 10× effective context and 89% more SWE-Milestone tasks resolved at 27% lower cost.
Install
$
npm install -g @spinejit/spine-codex@latestInstall command taken from the project's README or registry entry. Always review code before running it.
README excerpt
Read full README on GitHub →SpineCodex
Let your Codex work, evolve, and scale on the SpineTree.
·
English · 简体中文
Why SpineCodex
SpineCodex gives your Codex a SpineTree to work on: long-running, multi-step work is broken into owned Work Units, persisted by the runtime as SpineBranches, and refined, delegated, and completed as the tree evolves — without forcing the entire process into one ever-growing transcript.
Get started
Install it in your existing Codex environment and run it directly. The current
development branch targets upstream OpenAI Codex 0.153.4. Published builds
are listed in the release history:
npm install -g @spinejit/spine-codex@latest
spine-codex
Spine Spawn is enabled by default. Run /experimental to enable the optional
Memory Projection surface, then save and start a new conversation. Set
spine_spawn.max_concurrent_threads_per_session in ~/.codex/config.toml to
configure the total per-session thread limit, including the root thread.
What this enables
Compared with Codex, SpineCodex resolves 89% more tasks at 27% lower total cost on SWE-Milestone and extends the effective working context by up to 10×. It also improves the average score by 10.8 points on ProgramBench and the mean score by 9.2 points on FrontierSWE.
| Linear context | SpineCodex |
|---|---|
| ❌Run out of context? | ✅256K → 2.5M Effective Working ContextSpineJIT compiles completed branches into semantic Node Memory, extending effective working context beyond the native window. |
| ❌Drift after repeated compaction? | ✅**Minimum Effective Context. Maximum Focus.**Spine Runtime maintains the SpineTree and projects only the context required by the current Work Unit, keeping the agent focused. |
| ❌Lose patience and focus on long tasks? | ✅**Recursive Subagent Scaling on Demand.**SpineJIT lets the agent recursively unfold into specialized subagents on demand, bringing divide-and-conquer structure and greater reasoning depth to complex problems. |
What's new
Upcoming
- PI and DeepSeek Harness plugins — SpineSDK integrations in development.
- App Spine UI — Inspect and operate the SpineTree, Work Units, and runtime state from the Codex App. In development.
Preview App Spine UI
Preview — App Spine UI is in development.
Versions
0.4.0
Moves the upstream Codex baseline to 0.153.4. Unbinds the source-ledger
history cap from the one-turn visible item limit, and the 1MiB synthetic
context-plan cap, so long sessions are no longer killed at 4096 raw history
items or 1048576 bytes of Node Memory.
0.3.3
Similar plugs
Memori is agent-native memory infrastructure. A LLM-agnostic layer that turns agent execution and conversation into structured, persistent state for production systems. Built…
