Give your AI tools the context they’re missing. Store once, access from Claude, ChatGPT, Cursor, and everything else.
How it works
Three steps to context that works as hard as you do.
The missing layer
Docs, decisions, rules, examples — add once, reuse endlessly. Organize by project, tag for discovery, upload files and markdown. One searchable home for everything worth remembering.
One source, every tool
MCP connects your knowledge directly to Claude, ChatGPT, Cursor, and more. Use the CLI to power your agents. Hit the API for custom workflows. Your context, everywhere you work.
Gets better with use
Save valuable AI outputs back. Capture decisions as they happen. Your knowledge base grows with every interaction — better context in, better results out.
One context layer. Every AI tool. No more re-explaining.
Project wiki
Raw notes in, curated knowledge out. CTX.me maintains a structured, always-current wiki for every project — so your AI answers from organized knowledge, not a pile of files.
Summaries, entities, and concepts distilled from your docs, decisions, and files. You steer the structure; the wiki does the writing.
Ask a question, get a synthesized answer with sources you can click and verify. No black-box confidence.
Edit or delete a source and the wiki flags itself out of date. Regenerate on demand, or let it stay current automatically.
Conflicting facts across pages get flagged, not papered over. Your knowledge base stays trustworthy as it grows.
Available on Pro. Your raw context stays searchable on every plan.
I am
Push your architecture docs, conventions, and rules once. Every AI conversation starts with full context — whether you're in Claude Code, Cursor, or switching between models.
How it works
Nothing is sent to an external model unless you opt that project in. Not opted in? Search still works, entirely on our infrastructure.
Writes from AI tools are versioned and undoable, and every item records whether it came from you, your CLI, or an AI. Let agents contribute without holding your breath.
API keys are scoped (read / write / admin), stored only as hardened hashes, and never written to disk. Rotate or revoke any time.
Monthly AI usage allowances with a full ledger of every model call — what ran, where, and what it cost.
Start free, upgrade when you want the AI layer. Every plan includes full CLI, API, and MCP access.
For solo builders keeping context aligned.
For power users who want a self-maintaining knowledge base.
For heavy users who live in their context layer.