Skip to main content

​
Mental model

MemContext is infrastructure, not an application feature. Treat it the same way you treat a database or auth provider — a service your application depends on for a specific capability. Your application calls the MemContext API (or connects via MCP) to save and retrieve knowledge. It does not need to know about embeddings, classification, or version chains.

​
Integration pattern

A typical integration follows four steps:
  1. Save durable user or project context via POST /api/memories
  2. Search before generation or decision-making via GET /api/memories/search
  3. Profile (optional) — fetch broad context in one call via GET /api/memories/profile
  4. Feedback — record retrieval quality via POST /api/memories/:id/feedback

​
Example: Support bot

User: "I'm having trouble with my subscription"

Bot workflow:
1. search_memory({ query: "user subscription plan and billing" })
   → finds: "User is on the Pro plan, subscribed via Stripe"
   → finds: "User had a billing issue resolved on March 15"

2. Generate response with memory context injected into prompt

3. memory_feedback({ memoryId: "...", type: "helpful" })

​
Good fit use cases

Use caseMemory role
Coding assistantsStable preferences, project decisions, architecture context
Content generationBrand rules, evolving strategy, audience insights
Support botsCustomer history, past issues, plan details
CRM memoryRelationship context, interaction history
OnboardingUser progress, preferences learned during setup
PersonalizationStyle preferences, feature usage patterns

​
Ownership boundary

Keep your application logic outside MemContext. MemContext owns:
  • Storage and retrieval
  • Deduplication and classification
  • Version chains and history
  • Temporal filtering
  • Hybrid search
  • MCP tool access
Your application owns:
  • Business-specific prompts and logic
  • UI and user experience
  • Workflow orchestration
  • Downstream actions and integrations

​
API integration example

const MEMCONTEXT_API = "https://api.memcontext.in";

async function searchMemory(query: string, project?: string) {
  const params = new URLSearchParams({ query });
  if (project) params.set("project", project);

  const res = await fetch(`${MEMCONTEXT_API}/api/memories/search?${params}`, {
    headers: { "X-API-Key": process.env.MEMCONTEXT_API_KEY },
  });

  return res.json();
}

async function saveMemory(content: string, category: string, project?: string) {
  const res = await fetch(`${MEMCONTEXT_API}/api/memories`, {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "X-API-Key": process.env.MEMCONTEXT_API_KEY,
    },
    body: JSON.stringify({ content, category, project }),
  });

  return res.json();
}

​
Tips

  • Search before saving — avoid duplicates by checking if the knowledge already exists
  • Use projects to scope memories so different applications or contexts do not interfere
  • Category matters — preference and fact go into profile static context; everything goes into dynamic
  • Do not save noise — only persist information with long-term value