Memory Overview

Persistent memory for AI agents

The problem

Your AI agent forgets everything between conversations. Users repeat themselves, context is lost, and the experience feels broken.

The solution

Memory gives your agent persistent, searchable storage for facts extracted from conversations. When a user says “I prefer dark mode” or “The client is risk-averse,” those facts are stored and can be retrieved later.

How it works

1. Add memories - Pass conversation messages, we extract discrete facts
2. Search memories - Query by meaning, not just keywords
3. Filter by tags - Scope memories using 12 generic indexed tag fields

The API handles:

• Fact extraction (turning “I like X but hate Y” into separate facts)
• Deduplication (updating existing facts instead of creating duplicates)
• Semantic search (finding “investment preferences” when you stored “conservative strategies”)

Quick start

cURL

# Add a memory
curl -X POST https://api.case.dev/memory/v1 \
  -H "Authorization: Bearer $CASEDEV_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [
      {"role": "user", "content": "Remember that this client prefers conservative investments."}
    ],
    "tag_1": "client_123",
    "tag_2": "matter_456",
    "category": "preference"
  }'

# Search memories
curl -X POST https://api.case.dev/memory/v1/search \
  -H "Authorization: Bearer $CASEDEV_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "investment preferences",
    "tag_1": "client_123",
    "top_k": 5
  }'

Tag fields

Memory provides 12 generic indexed tag fields (tag_1 through tag_12) for filtering. You decide what each tag means for your application:

Use Case	tag_1	tag_2	tag_3	tag_4
Legal	client_id	matter_id	attorney_id	session_id
Healthcare	patient_id	encounter_id	provider_id	department
E-commerce	customer_id	order_id	product_id	session_id
Support	user_id	ticket_id	agent_id	channel

All tag fields are indexed for fast filtering. Use them for any dimension you need to query by.

Example: Legal application

cURL

# Store a memory scoped to client + matter
curl -X POST https://api.case.dev/memory/v1 \
  -H "Authorization: Bearer $CASEDEV_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [
      {"role": "user", "content": "The patent filing deadline is March 15th, 2026."}
    ],
    "tag_1": "client_acme_corp",
    "tag_2": "matter_patent_2026_001",
    "tag_3": "atty_jane_smith",
    "category": "deadline"
  }'

# Search only this client's deadlines
curl -X POST https://api.case.dev/memory/v1/search \
  -H "Authorization: Bearer $CASEDEV_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "upcoming deadlines",
    "tag_1": "client_acme_corp",
    "category": "deadline",
    "top_k": 10
  }'

# List all memories for a specific matter
curl "https://api.case.dev/memory/v1?tag_1=client_acme_corp&tag_2=matter_patent_2026_001" \
  -H "Authorization: Bearer $CASEDEV_API_KEY"

Categories

The category field is free-form text - you define your own categories. Here are suggested categories for different domains:

Legal categories

Category	Description	Example
fact	General facts about client/matter	”Client founded company in 2019”
preference	Communication or work preferences	”Client prefers email updates”
deadline	Filing deadlines, court dates	”Motion due March 15th”
decision	Decisions made by client/court	”Client approved settlement terms”
contact	Contact information	”Client’s assistant is Jane ([e-mail])“
relationship	Relationships between parties	”Defendant is client’s former business partner”
court_ruling	Court rulings and precedents	”Judge denied motion to dismiss”
procedural	Procedural information	”Case assigned to Judge Thompson”

Healthcare categories

Category	Description
allergy	Patient allergies
medication	Current medications
preference	Treatment preferences
history	Medical history
family_history	Family medical history
social	Social/lifestyle factors

Filtering by category

cURL

# Get only deadline memories for a client
curl "https://api.case.dev/memory/v1?tag_1=client_123&category=deadline" \
  -H "Authorization: Bearer $CASEDEV_API_KEY"

# Search preferences only
curl -X POST https://api.case.dev/memory/v1/search \
  -H "Authorization: Bearer $CASEDEV_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "communication method",
    "tag_1": "client_123",
    "category": "preference"
  }'

Listing memories

cURL

# List all memories for a client
curl "https://api.case.dev/memory/v1?tag_1=client_123&limit=50" \
  -H "Authorization: Bearer $CASEDEV_API_KEY"

# List with multiple filters
curl "https://api.case.dev/memory/v1?tag_1=client_123&tag_2=matter_456&category=deadline&limit=20&offset=0" \
  -H "Authorization: Bearer $CASEDEV_API_KEY"

Deleting memories

Delete a single memory

cURL

curl -X DELETE https://api.case.dev/memory/v1/mem_abc123 \
  -H "Authorization: Bearer $CASEDEV_API_KEY"

Bulk delete by tags

cURL

curl -X POST https://api.case.dev/memory/v1/delete-many \
  -H "Authorization: Bearer $CASEDEV_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{}'

Bulk delete removes all matching memories. Use with caution.

Memory events

When you add memories, the API returns what happened to each extracted fact:

Event	Description
ADD	New memory created
UPDATE	Existing memory was updated/merged
DELETE	Contradicting info caused memory deletion
NONE	Duplicate - memory already exists

cURL

curl -X POST https://api.case.dev/memory/v1/add \
  -H "Authorization: Bearer $CASEDEV_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{}'

Direct storage (skip extraction)

To store a fact directly without LLM extraction, set infer: false:

cURL

curl -X POST https://api.case.dev/memory/v1/add \
  -H "Authorization: Bearer $CASEDEV_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{}'

Custom extraction prompts

For domain-specific extraction, provide a custom prompt:

cURL

curl -X POST https://api.case.dev/memory/v1/add \
  -H "Authorization: Bearer $CASEDEV_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{}'

Pricing

Operation	Cost
Add memory	$0.001 per operation
Search	$0.001 per operation
List/Get/Delete	Free

LLM usage for fact extraction and embeddings is billed separately through the LLM service at standard rates.

Related services

LLMs

Use memories as context for chat completions

Vault

Store documents, use Memory for extracted facts