
VaultCrux Memory Core
Give long-running coding agents durable memory for knowledge, decisions, constraints, signals, and coverage via VaultCrux’s 32-tool Memory Core API.
Overview
VaultCrux Memory Core is a Build-phase MCP server that provides 32 tools for agent memory across knowledge, decisions, constraints, signals, and coverage.
What is this MCP server?
- 32 MCP tools spanning knowledge, decisions, constraints, signals, and coverage
- Remote streamable-http endpoint at api.vaultcrux.com/memory-core/mcp
- Tenant-scoped reads with x-api-key and required x-tenant-id headers
- Optional x-agent-id for query and claim-check context
- VaultCrux Memory Core server version 0.2.0 from CueCrux/VaultCrux repo
- 32 MCP tools documented in server description
- Remote endpoint: api.vaultcrux.com/memory-core/mcp
What problem does it solve?
Agents forget prior decisions and constraints every session, so you re-explain architecture rules and repeat mistakes.
Who is it for?
Solo builders running multi-session AI dev agents who need tenant-scoped, decision-aware memory on VaultCrux.
Skip if: Simple one-off scripts with no persistence needs, or teams unwilling to use a hosted VaultCrux tenant.
What do I get? / Deliverables
After connecting with API key and tenant headers, your agent can query and update structured VaultCrux memory through 32 dedicated MCP tools.
- Remote MCP connection to Memory Core with tenant-scoped memory reads
- Access to 32 tools for knowledge, decisions, constraints, signals, and coverage
Recommended MCP Servers
Journey fit
Agent memory infrastructure is built while you shape how your product’s AI helpers remember context, not after launch analytics alone. Agent-tooling is the canonical shelf: Memory Core is purpose-built MCP access to structured agent memory primitives.
How it compares
Hosted agent-memory MCP (32 tools), not a local filesystem skill or generic embedding search plugin.
Common Questions / FAQ
Who is VaultCrux Memory Core for?
Developers building agent-heavy products who want VaultCrux to hold knowledge, decisions, constraints, signals, and coverage accessible over MCP.
When should I use VaultCrux Memory Core?
Use it during Build while designing agent workflows that must remember constraints and decisions across sessions and scoped tenants.
How do I add VaultCrux Memory Core to my agent?
Register the remote MCP URL https://api.vaultcrux.com/memory-core/mcp with streamable-http transport and set x-api-key, x-tenant-id, and optionally x-agent-id in your client.