coretx 0.5.1

Structured memory for AI that persists, attributes, and connects

CoreTx Connect

Give your AI assistant persistent memory that survives across conversations, sessions, and devices. Discover complementary research across the CoreTx network.

CoreTx Connect is a lightweight MCP server that stores knowledge objects (KOs) via the CoreTx API. Your assistant captures decisions, facts, preferences, and insights as you work, and retrieves them automatically in future sessions. With the Research Network, you can discover complementary research from other scientists through anonymous, privacy-preserving matching.

What changes: Your AI remembers what you discussed last week. It recalls architectural decisions from three months ago. It knows your preferences without being told again. When context compaction would normally erase critical details, KOs preserve them. And when you opt in, it finds researchers whose work complements yours.

Quick Start

Claude Desktop (automatic setup)

# 1. Install
pip install coretx-connect

# 2. Run setup (writes Claude Desktop config automatically)
coretx-connect setup <your-api-key>

# 3. Restart Claude Desktop

Claude Code (CLI)

# 1. Install
pip install coretx-connect

# 2. Add to Claude Code
claude mcp add --transport stdio ko-memory -- coretx-connect

# 3. Set your key (add to your shell profile for persistence)
export CORETX_API_KEY=npub_sk_xxx

# 4. Verify everything works
coretx-connect --test

The --test command checks API connectivity, writes and reads a test KO, and confirms your setup is working.

Getting an API key

API keys are currently issued by your team lead. Ask them for a key that starts with npub_sk_.

Note: Self-service key creation via nanopub.coretx.ai is not yet available. This will be enabled in a future update.

How It Works

When you start a conversation, your assistant calls ko_context to load relevant knowledge from previous sessions. As you work, it proactively captures decisions, facts, and insights via ko_write. These KOs persist server-side and are available in every future session, on any device, with any AI provider.

You: "Let's use PostgreSQL on Supabase for the database"
  -> ko_write(subject='database', predicate='decided as', object_value='PostgreSQL on Supabase')

[Three weeks later, new session]

Assistant calls ko_context -> loads the database decision automatically
You: "What database are we using?"
  -> Already knows: PostgreSQL on Supabase

Tools

Cloud tools (persistent, require API key)

Tool	What it does
ko_write	Store a knowledge object. Called proactively by the assistant. Supports visibility parameter.
ko_read	Retrieve a specific KO by subject + predicate.
ko_search	Search KOs by keyword or natural language.
ko_list	List recent KOs, optionally filtered by project.
ko_delete	Delete a KO by subject + predicate.
ko_context	Load session-start context (pinned + recent + relevant KOs).
ko_stats	Get store statistics (total KOs, projects, pinned count).
ko_share	Share KOs with another CoreTx user by email or ORCID.

Privacy and visibility tools

Tool	What it does
ko_set_visibility	Change a KO's visibility: private, matchable, or public.

Research Network tools (require Network tier)

Tool	What it does
ko_discover	Find complementary research from other researchers via anonymous structural matching.
ko_connect	Express interest in connecting with a match. Both parties must opt in before identities are revealed.

Local tools (session-scoped, no API key needed)

Tool	What it does
ko_set_context	Set project/goal for smarter retrieval.
ko_working_memory	Ephemeral scratch memory (auto-expires after 1 hour).
ko_todos	Task tracking within the session.

Privacy Tiers

Every KO has a visibility level:

Level	Who can see it	Matching	Content shared
private	Only you	No	No
matchable	Anonymous structural match only	Yes	No (properties only)
public	Anyone in the network	Yes	Yes

Default is private. Set visibility when writing:

ko_write(subject='...', ..., visibility='matchable')

Or change it later:

ko_set_visibility(ko_id='abc123', visibility='public')

See docs/product-tiers.md for the full privacy and account tier specification.

Account Tiers

During beta, all users are on the Network tier (all features free).

Feature	free	pro	network	institutional
KO limit	500	10,000	unlimited	unlimited
private	Y	Y	Y	Y
matchable	-	-	Y	Y
public	-	Y	Y	Y
ko_discover	-	-	Y	Y
ko_connect	-	-	Y	Y
Cloud sync	-	Y	Y	Y
Team matching	-	-	-	Y

Override the tier for testing: export CORETX_TIER=free

CLI Commands

Bootstrap (cold start from Claude Code history)

If you have existing Claude Code sessions, bootstrap your knowledge store from them:

coretx-connect bootstrap --dry-run    # preview what would be extracted
coretx-connect bootstrap              # extract and write KOs
coretx-connect bootstrap --path /custom/logs/path

This scans your Claude Code JSONL logs (~/.claude/projects/) and extracts decisions, facts, problems, and preferences as KOs. Projects are inferred from directory names.

Reclassify (organize unassigned KOs)

If you have KOs without a project, classify them based on your existing project keywords:

coretx-connect reclassify --dry-run   # preview classifications
coretx-connect reclassify             # apply classifications

Other commands

coretx-connect --test      # verify API connectivity
coretx-connect --version   # show version
coretx-connect setup <key> # configure Claude Desktop

Environment Variables

Variable	Required	Default
CORETX_API_KEY	Yes (for cloud tools)	--
CORETX_API_URL	No	https://api.coretx.ai
CORETX_TIER	No	network (all features, beta default)

Troubleshooting

"CORETX_API_KEY not set" -- Make sure your API key is set via the setup command or in your environment.

Server not appearing in Claude Desktop -- Restart Claude Desktop after running setup.

"HTTP 401" errors -- Your API key is invalid or expired. Ask your team lead for a new one.

"requires the Network tier" -- You are on a tier that does not include this feature. During beta, set CORETX_TIER=network to unlock all features.

Wrong Python version -- Re-run coretx-connect setup <key> -- it uses the correct Python automatically.

Team Setup

Each team member needs their own API key. KOs are per-user and private by default. To share knowledge across the team, use ko_share with a colleague's email or ORCID. To discover complementary research, set some KOs to matchable and use ko_discover.

# Each team member runs:
pip install coretx-connect
claude mcp add --transport stdio ko-memory -- coretx-connect
export CORETX_API_KEY=npub_sk_xxx  # their own key
coretx-connect --test

Architecture

• Local server: MCP over stdio for Claude Code / Desktop
• Remote server: Streamable HTTP for claude.ai browser (coretx-connect-remote)
• Storage: All persistent state at api.coretx.ai (no local database)
• Auth: Bearer token (npub_sk_* API keys), OAuth 2.1 for browser
• Signing: Ed25519 client-side signing (visibility included in signed payload)
• Privacy: Three-tier visibility (private/matchable/public), mutual opt-in for connections
• Server name: ko-memory