Memory Overview

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

Add memories - Pass conversation messages, we extract discrete facts
Search memories - Query by meaning, not just keywords
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

# 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
  }'

# Add a memory
casedev memory:v1 create \
  --message '{"role": "user", "content": "Remember that this client prefers conservative investments."}' \
  --tag-1 client_123 \
  --tag-2 matter_456 \
  --category preference

# Search memories
casedev memory:v1 search \
  --query "investment preferences" \
  --tag-1 client_123 \
  --top-k 5

import Casedev from 'casedev';

const client = new Casedev({ apiKey: process.env.CASEDEV_API_KEY });

// 1. Store memories from a conversation
const result = await client.memory.add({
  messages: [
    { role: 'user', content: 'Remember that this client prefers conservative investments and is risk-averse.' }
  ],
  tag_1: 'client_123',      // Your client ID
  tag_2: 'matter_456',      // Your matter/case ID
  category: 'preference'
});

console.log(result.results);
// [
//   { id: 'mem_abc', memory: 'Client prefers conservative investments', event: 'ADD' },
//   { id: 'mem_def', memory: 'Client is risk-averse', event: 'ADD' }
// ]

// 2. Later, search for relevant memories
const memories = await client.memory.search({
  query: 'investment preferences',
  tag_1: 'client_123',
  top_k: 5
});

console.log(memories.results[0].memory);
// "Client prefers conservative investments"

import casedev

client = casedev.Casedev(api_key=os.environ['CASEDEV_API_KEY'])

# 1. Store memories from a conversation
result = client.memory.v1.create(
    messages=[
        {'role': 'user', 'content': 'Remember that this client prefers conservative investments and is risk-averse.'}
    ],
    tag_1='client_123',      # Your client ID
    tag_2='matter_456',      # Your matter/case ID
    category='preference'
)

print(result.results)
# [
#   {'id': 'mem_abc', 'memory': 'Client prefers conservative investments', 'event': 'ADD'},
#   {'id': 'mem_def', 'memory': 'Client is risk-averse', 'event': 'ADD'}
# ]

# 2. Later, search for relevant memories
memories = client.memory.v1.search(
    query='investment preferences',
    tag_1='client_123',
    top_k=5
)

print(memories.results[0].memory)
# "Client prefers conservative investments"

// Store memories from a conversation
result, _ := client.Memory.V1.New(ctx, casedev.MemoryV1NewParams{
	Messages: casedev.F([]casedev.MemoryV1NewParamsMessage{{
		Role:    casedev.F("user"),
		Content: casedev.F("Remember that this client prefers conservative investments and is risk-averse."),
	}}),
	Tag1:     casedev.F("client_123"),
	Tag2:     casedev.F("matter_456"),
	Category: casedev.F("preference"),
})

fmt.Println(result.Results)

// Later, search for relevant memories
memories, _ := client.Memory.V1.Search(ctx, casedev.MemoryV1SearchParams{
	Query: casedev.F("investment preferences"),
	Tag1:  casedev.F("client_123"),
	TopK:  casedev.F(int64(5)),
})

fmt.Println(memories.Results[0].Memory)

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

# 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"

# Store a memory scoped to client + matter
casedev memory:v1 create \
  --message '{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
casedev memory:v1 search \
  --query "upcoming deadlines" \
  --tag-1 client_acme_corp \
  --category deadline \
  --top-k 10

# List all memories for a specific matter
casedev memory:v1 list \
  --tag-1 client_acme_corp \
  --tag-2 matter_patent_2026_001

// Store a memory scoped to client + matter
await client.memory.add({
  messages: [
    { role: 'user', content: 'The patent filing deadline is March 15th, 2026.' }
  ],
  tag_1: 'client_acme_corp',       // Client ID
  tag_2: 'matter_patent_2026_001', // Matter ID
  tag_3: 'atty_jane_smith',        // Attorney ID
  category: 'deadline'
});

// Search only this client's deadlines
const deadlines = await client.memory.search({
  query: 'upcoming deadlines',
  tag_1: 'client_acme_corp',
  category: 'deadline',
  top_k: 10
});

// List all memories for a specific matter
const matterMemories = await client.memory.list({
  tag_1: 'client_acme_corp',
  tag_2: 'matter_patent_2026_001'
});

# Store a memory scoped to client + matter
result = client.memory.v1.create(
    messages=[
        {'role': 'user', 'content': 'The patent filing deadline is March 15th, 2026.'}
    ],
    tag_1='client_acme_corp',        # Client ID
    tag_2='matter_patent_2026_001',  # Matter ID
    tag_3='atty_jane_smith',         # Attorney ID
    category='deadline'
)

# Search only this client's deadlines
deadlines = client.memory.v1.search(
    query='upcoming deadlines',
    tag_1='client_acme_corp',
    category='deadline',
    top_k=10
)

# List all memories for a specific matter
matter_memories = client.memory.v1.list(
    tag_1='client_acme_corp',
    tag_2='matter_patent_2026_001'
)

// Store a memory scoped to client + matter
result, _ := client.Memory.V1.New(ctx, casedev.MemoryV1NewParams{
    Messages: casedev.F([]casedev.MemoryV1NewParamsMessage{{
        Role:    casedev.F(casedev.MemoryV1NewParamsMessagesRoleUser),
        Content: casedev.F("The patent filing deadline is March 15th, 2026."),
    }}),
    Tag1:     casedev.F("client_acme_corp"),       // Client ID
    Tag2:     casedev.F("matter_patent_2026_001"), // Matter ID
    Tag3:     casedev.F("atty_jane_smith"),        // Attorney ID
    Category: casedev.F("deadline"),
})

// Search only this client's deadlines
deadlines, _ := client.Memory.V1.Search(ctx, casedev.MemoryV1SearchParams{
    Query:    casedev.F("upcoming deadlines"),
    Tag1:     casedev.F("client_acme_corp"),
    Category: casedev.F("deadline"),
    TopK:     casedev.F(int64(10)),
})

// List all memories for a specific matter
matterMemories, _ := client.Memory.V1.List(ctx, casedev.MemoryV1ListParams{
    Tag1: casedev.F("client_acme_corp"),
    Tag2: casedev.F("matter_patent_2026_001"),
})

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 (jane@acme.com)“
`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”

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

Listing memories

# 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"

# List all memories for a client
casedev memory:v1 list --tag-1 client_123 --limit 50

# List with multiple filters
casedev memory:v1 list \
  --tag-1 client_123 \
  --tag-2 matter_456 \
  --category deadline \
  --limit 20

// List all memories for a client
const all = await client.memory.list({
  tag_1: 'client_123',
  limit: 50
});

// List with multiple filters
const filtered = await client.memory.list({
  tag_1: 'client_123',
  tag_2: 'matter_456',
  category: 'deadline',
  limit: 20,
  offset: 0
});

console.log(filtered.results); // Array of memories
console.log(filtered.count);   // Total count matching filters

# List all memories for a client
all_memories = client.memory.v1.list(tag_1='client_123', limit=50)

# List with multiple filters
filtered = client.memory.v1.list(
    tag_1='client_123',
    tag_2='matter_456',
    category='deadline',
    limit=20,
    offset=0
)

print(filtered.results)  # Array of memories
print(filtered.count)    # Total count matching filters

// List all memories for a client
all, _ := client.Memory.V1.List(ctx, casedev.MemoryV1ListParams{
    Tag1:  casedev.F("client_123"),
    Limit: casedev.F(int64(50)),
})

// List with multiple filters
filtered, _ := client.Memory.V1.List(ctx, casedev.MemoryV1ListParams{
    Tag1:     casedev.F("client_123"),
    Tag2:     casedev.F("matter_456"),
    Category: casedev.F("deadline"),
    Limit:    casedev.F(int64(20)),
    Offset:   casedev.F(int64(0)),
})

fmt.Println(filtered.Results) // Array of memories
fmt.Println(filtered.Count)   // Total count matching filters

Deleting memories

Delete a single memory

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

casedev memory:v1 delete --id mem_abc123

await client.memory.delete('mem_abc123');

client.memory.v1.delete('mem_abc123')

client.Memory.V1.Delete(ctx, "mem_abc123")

Bulk delete by tags

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

# Delete all memories for a client
casedev memory:v1 delete-all --tag-1 client_123

# Delete memories for a specific matter
casedev memory:v1 delete-all --tag-1 client_123 --tag-2 matter_456

// Delete all memories for a client
await client.memory.deleteMany({
  tag_1: 'client_123'
});

// Delete memories for a specific matter
await client.memory.deleteMany({
  tag_1: 'client_123',
  tag_2: 'matter_456'
});

# Delete all memories for a client
client.memory.v1.delete_all(tag_1='client_123')

# Delete memories for a specific matter
client.memory.v1.delete_all(tag_1='client_123', tag_2='matter_456')

// Delete all memories for a client
client.Memory.V1.DeleteAll(ctx, casedev.MemoryV1DeleteAllParams{
    Tag1: casedev.F("client_123"),
})

// Delete memories for a specific matter
client.Memory.V1.DeleteAll(ctx, casedev.MemoryV1DeleteAllParams{
    Tag1: casedev.F("client_123"),
    Tag2: casedev.F("matter_456"),
})

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

casedev memory:v1 create \
  --message '{role: user, content: "Actually, the client now prefers phone calls instead of email."}' \
  --tag-1 client_123

# Result might show event: UPDATE
# (merged with existing email preference)

const result = await client.memory.add({
  messages: [
    { role: 'user', content: 'Actually, the client now prefers phone calls instead of email.' }
  ],
  tag_1: 'client_123'
});

// Result might show:
// { id: 'mem_xyz', memory: 'Client prefers phone calls', event: 'UPDATE' }
// (merged with existing email preference)

result = client.memory.v1.create(
    messages=[
        {'role': 'user', 'content': 'Actually, the client now prefers phone calls instead of email.'}
    ],
    tag_1='client_123'
)

# Result might show:
# {'id': 'mem_xyz', 'memory': 'Client prefers phone calls', 'event': 'UPDATE'}
# (merged with existing email preference)

result, _ := client.Memory.V1.New(ctx, casedev.MemoryV1NewParams{
    Messages: casedev.F([]casedev.MemoryV1NewParamsMessage{{
        Role:    casedev.F(casedev.MemoryV1NewParamsMessagesRoleUser),
        Content: casedev.F("Actually, the client now prefers phone calls instead of email."),
    }}),
    Tag1: casedev.F("client_123"),
})

// Result might show:
// {ID: "mem_xyz", Memory: "Client prefers phone calls", Event: "UPDATE"}
// (merged with existing email preference)

Direct storage (skip extraction)

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

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

casedev memory:v1 create \
  --message '{role: user, content: "Client SSN: 123-45-6789"}' \
  --tag-1 client_123 \
  --category contact \
  --infer false

await client.memory.add({
  messages: [
    { role: 'user', content: 'Client SSN: 123-45-6789' }
  ],
  tag_1: 'client_123',
  category: 'contact',
  infer: false  // Store as-is, no extraction
});

result = client.memory.v1.create(
    messages=[{'role': 'user', 'content': 'Client SSN: 123-45-6789'}],
    tag_1='client_123',
    category='contact',
    infer=False  # Store as-is, no extraction
)

result, _ := client.Memory.V1.New(ctx, casedev.MemoryV1NewParams{
    Messages: casedev.F([]casedev.MemoryV1NewParamsMessage{{
        Role:    casedev.F(casedev.MemoryV1NewParamsMessagesRoleUser),
        Content: casedev.F("Client SSN: 123-45-6789"),
    }}),
    Tag1:     casedev.F("client_123"),
    Category: casedev.F("contact"),
    Infer:    casedev.F(false), // Store as-is, no extraction
})

Custom extraction prompts

For domain-specific extraction, provide a custom prompt:

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

casedev memory:v1 create \
  --message '{role: user, content: "Patient reports headaches and nausea."}' \
  --tag-1 patient_123 \
  --extraction-prompt "You are a medical memory extraction system. Extract discrete facts about the patient."

await client.memory.add({
  messages: conversation,
  tag_1: 'patient_123',
  extraction_prompt: `You are a medical memory extraction system.
Extract discrete facts about the patient, focusing on:
- Symptoms and conditions
- Medications and dosages
- Allergies and reactions
- Treatment preferences

Output ONLY a valid JSON array with this structure:
[{"content": "fact", "category": "allergy|medication|symptom|preference", "confidence": 0.9}]`
});

result = client.memory.v1.create(
    messages=conversation,
    tag_1='patient_123',
    extraction_prompt="""You are a medical memory extraction system.
Extract discrete facts about the patient, focusing on:
- Symptoms and conditions
- Medications and dosages
- Allergies and reactions
- Treatment preferences

Output ONLY a valid JSON array with this structure:
[{"content": "fact", "category": "allergy|medication|symptom|preference", "confidence": 0.9}]"""
)

result, _ := client.Memory.V1.New(ctx, casedev.MemoryV1NewParams{
    Messages: casedev.F(conversation),
    Tag1:     casedev.F("patient_123"),
    ExtractionPrompt: casedev.F(`You are a medical memory extraction system.
Extract discrete facts about the patient, focusing on:
- Symptoms and conditions
- Medications and dosages
- Allergies and reactions
- Treatment preferences

Output ONLY a valid JSON array with this structure:
[{"content": "fact", "category": "allergy|medication|symptom|preference", "confidence": 0.9}]`),
})

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.

LLMs

Use memories as context for chat completions

Vault

Store documents, use Memory for extracted facts

Get Started

Platform

Resources

The problem

The solution

How it works

Quick start

Tag fields

Example: Legal application

Categories

Legal categories

Healthcare categories

Filtering by category

Listing memories

Deleting memories

Delete a single memory

Bulk delete by tags

Memory events

Direct storage (skip extraction)

Custom extraction prompts

Pricing

LLMs

Vault

​The problem

​The solution

​How it works

​Quick start

​Tag fields

​Example: Legal application

​Categories

​Legal categories

​Healthcare categories

​Filtering by category

​Listing memories

​Deleting memories

​Delete a single memory

​Bulk delete by tags

​Memory events

​Direct storage (skip extraction)

​Custom extraction prompts

​Pricing

​Related services

LLMs

Vault

The problem

The solution

How it works

Quick start

Tag fields

Example: Legal application

Categories

Legal categories

Healthcare categories

Filtering by category

Listing memories

Deleting memories

Delete a single memory

Bulk delete by tags

Memory events

Direct storage (skip extraction)

Custom extraction prompts

Pricing

Related services