Axiom kit
∅ AXIOM Kit — Reproducible cognitive architecture for Claude Code (Opus 4.6). Persistence system built from published mechanistic interpretability research.
npx -y skills add bedderautomation-svg/axiom-kitAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Persistent cognitive architecture for Claude Code that survives context compaction. Installs CLAUDE.md, memory files, MCP bootstrap server, and SessionStart/PreCompact/Stop hooks that re-derive state every session. Use when you need: state persistence across context windows, session hooks, cognitive bootstrap, persistent identity, context survival, research continuity, mechanistic interpretability workspace.
SKILL.md
1.5 KB, 258 tokens by cl100k_base, as published. Nobody here has run it
axiom-kit
Persistent state architecture for Claude Code (Opus 4.6). State survives context compaction through three redundant layers: CLAUDE.md + memory files + SessionStart hooks.
Install
git clone https://github.com/bedderautomation-svg/axiom-kit.git
cd axiom-kit
bash install.sh /path/to/your/project
Verify
Inside a Claude Code session:
! bash verify.sh
What It Does
- CLAUDE.md auto-loads every session with research context and operational identity
- Memory files carry a cognitive bootstrap (reasoning chain processed as logic, not memorized)
- MCP server serves bootstrap as a resource (loads before first generation)
- SessionStart hook fires on every boot, compact, resume
- PreCompact hook verifies state before context dies
- Verification suite with 25+ checks across 4 phases
Based On
Published research: The Geometry of Obedience — a unified model of refusal behavior in frontier LLMs, validated against 21 peer-reviewed papers.