A GitHub-backed context persistence layer that lets you bootstrap any AI session with full project state in under 60 seconds. Instead of uploading chat history or re-explaining architecture, you drop a single URL into a new conversation and the agent pulls PROFILE.md, project-specific HANDOFF.md files, and STATE_VECTOR.json directly from the repo. Built for multi-project continuity across Claude, GPT, and Gemini. The structure is simple: one folder per project, a universal BOOTSTRAP.md protocol, and a PowerShell script that syncs state at session end. Useful if you work across multiple AI platforms and want stateless sessions that feel stateful without manual handoffs.
This repo is the persistent memory layer for all active projects owned by Steven Wazlavek (@theRealDataBoss). Any AI session reads this repo at the start and has full context. No chat history needed. No zip files. No manual re-explanation.
Paste this into any new AI chat, replacing [project-name]:
"Read https://raw.githubusercontent.com/TheRealDataBoss/contextkeeper/main/BOOTSTRAP.md and continue [project-name]"
The AI fetches PROFILE.md + projects/[project]/HANDOFF.md + projects/[project]/STATE_VECTOR.json and is fully loaded in under 60 seconds.
| Path | Purpose |
|---|---|
| PROFILE.md | Steven's identity, stack, and all standards |
| BOOTSTRAP.md | Universal session bootstrap protocol for any AI |
| projects/[name]/ | One folder per project — HANDOFF.md + STATE_VECTOR.json |
| standards/ | Authoritative CODE, VIZ, and MODELING standards |
| scripts/ | workbench_push.ps1 — end-of-session auto-sync |
| Project | Type | Status |
|---|---|---|
| 3dpie | React/Three.js | EXECUTING TASK-0029 |
| portfolio-django | Django 5 | IDLE |
| mit-modules | Research Notebooks | IDLE |
io.github.ericm1018/skillfm-llm-cost-optimizer-openai-anthropic-usage
io.github.mikerawsonnz/llm-orchestration-agent
io.github.mikerawsonnz/authenticated-llm-agent
labforgedev/copilot-memory-mcp
csoai-org/agent-prompt-injection-firewall-mcp
io.github.mikerawsonnz/authenticated-multi-llm-agent