> ## Documentation Index
> Fetch the complete documentation index at: https://docs.minns.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Get agent memories

> Retrieve long-term memories for a specific agent.

## Path parameters

<ParamField path="agent_id" type="integer" required>
  Agent identifier (`u64`).
</ParamField>

## Query parameters

<ParamField query="limit" type="integer" default="10">
  Maximum number of memories to return. Max: `1000`.
</ParamField>

## Response

Returns an array of `MemoryResponse` objects.

<ResponseField name="id" type="integer">
  Memory identifier.
</ResponseField>

<ResponseField name="agent_id" type="integer">
  Agent that formed the memory.
</ResponseField>

<ResponseField name="session_id" type="integer">
  Session identifier.
</ResponseField>

<ResponseField name="strength" type="number">
  Memory strength (`0.0`–`1.0`).
</ResponseField>

<ResponseField name="relevance_score" type="number">
  Relevance score (`0.0`–`1.0`).
</ResponseField>

<ResponseField name="access_count" type="integer">
  Number of times the memory has been accessed.
</ResponseField>

<ResponseField name="formed_at" type="integer">
  Formation timestamp (nanoseconds since Unix epoch).
</ResponseField>

<ResponseField name="last_accessed" type="integer">
  Last access timestamp.
</ResponseField>

<ResponseField name="context_hash" type="integer">
  Context fingerprint.
</ResponseField>

<ResponseField name="context" type="EventContext">
  Full context object. See [Types](/api-reference/types#eventcontext-structure).
</ResponseField>

<ResponseField name="outcome" type="string">
  Episode outcome description.
</ResponseField>

<ResponseField name="memory_type" type="string">
  One of: `"Episodic"`, `"Working"`, `"Semantic"`, `"Negative"`.
</ResponseField>

```json theme={null}
[
  {
    "id": 789,
    "agent_id": 1,
    "session_id": 5001,
    "strength": 0.92,
    "relevance_score": 0.88,
    "access_count": 3,
    "formed_at": 1738425600000000000,
    "last_accessed": 1738512000000000000,
    "context_hash": 12345678901234,
    "context": { "..." : "..." },
    "outcome": "Success",
    "memory_type": "Episodic"
  }
]
```