Projects

Projects let you scope memories, decisions, and patterns to specific codebases or initiatives. Each project has a unique slug used as a human-readable identifier throughout the API.

Register Project

POST /v1/projects

Request Body

Field	Type	Required	Description
name	string	Yes	Human-readable project name (1-255 characters)
slug	string	Yes	URL-safe identifier, lowercase alphanumeric + hyphens (1-100 characters, e.g. my-api)
description	string	No	Project description (max 5,000 characters)
stack	object	No	Technical stack metadata (max 10KB). Example: {"languages": ["python"], "frameworks": ["fastapi"]}
repo_url	string	No	Repository URL (max 500 characters)
metadata	object	No	Custom metadata (max 10KB)

Response 201 Created

{
  "id": "c1d2e3f4-a5b6-7890-cdef-123456789012",
  "slug": "my-api",
  "name": "My API Service",
  "description": "Production REST API for memory management",
  "stack": {
    "languages": ["python"],
    "frameworks": ["fastapi"],
    "databases": ["postgresql"]
  },
  "repo_url": "https://github.com/org/my-api",
  "metadata": {},
  "memory_count": 0,
  "created_at": "2026-03-17T10:00:00Z",
  "updated_at": "2026-03-17T10:00:00Z"
}

curl

curl -X POST https://api.memoryrelay.net/v1/projects \
  -H "Authorization: Bearer $MEMORYRELAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "slug": "my-api",
    "name": "My API Service",
    "description": "Production REST API for memory management",
    "stack": {
      "languages": ["python"],
      "frameworks": ["fastapi"],
      "databases": ["postgresql"]
    },
    "repo_url": "https://github.com/org/my-api"
  }'

Python SDK

from memoryrelay import MemoryRelay

client = MemoryRelay(api_key="mem_prod_...")

project = client.projects.create(
    slug="my-api",
    name="My API Service",
    description="Production REST API for memory management",
    stack={"languages": ["python"], "frameworks": ["fastapi"]},
    repo_url="https://github.com/org/my-api",
)
print(project.id)

---

List Projects

GET /v1/projects

Query Parameters

Parameter	Type	Default	Description
limit	integer	20	Results per page (1-100)
offset	integer	0	Number of results to skip

Response

{
  "data": [
    {
      "id": "c1d2e3f4-a5b6-7890-cdef-123456789012",
      "slug": "my-api",
      "name": "My API Service",
      "description": "Production REST API for memory management",
      "stack": {
        "languages": ["python"],
        "frameworks": ["fastapi"]
      },
      "repo_url": "https://github.com/org/my-api",
      "metadata": {},
      "memory_count": 42,
      "created_at": "2026-03-17T10:00:00Z",
      "updated_at": "2026-03-17T10:00:00Z"
    }
  ],
  "next_cursor": null,
  "has_more": false,
  "total_count": 1
}

curl

curl "https://api.memoryrelay.net/v1/projects?limit=20" \
  -H "Authorization: Bearer $MEMORYRELAY_API_KEY"

Python SDK

projects = client.projects.list()
for p in projects.data:
    print(f"{p.slug}: {p.name} ({p.memory_count} memories)")

---

Get Project

GET /v1/projects/{id}

Path Parameters

Parameter	Type	Description
id	UUID	Project ID

Response

Returns a single project object.

curl

curl https://api.memoryrelay.net/v1/projects/c1d2e3f4-a5b6-7890-cdef-123456789012 \
  -H "Authorization: Bearer $MEMORYRELAY_API_KEY"

Python SDK

project = client.projects.get("c1d2e3f4-a5b6-7890-cdef-123456789012")
print(f"{project.name}: {project.description}")

---

Update Project

PATCH /v1/projects/{id}

Path Parameters

Parameter	Type	Description
id	UUID	Project ID

Request Body

Field	Type	Required	Description
name	string	No	Updated name (max 255 characters)
description	string	No	Updated description (max 5,000 characters)
stack	object	No	Updated tech stack metadata
repo_url	string	No	Updated repository URL
metadata	object	No	Updated metadata

Response

Returns the updated project object.

curl

curl -X PATCH https://api.memoryrelay.net/v1/projects/c1d2e3f4-a5b6-7890-cdef-123456789012 \
  -H "Authorization: Bearer $MEMORYRELAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "stack": {
      "languages": ["python"],
      "frameworks": ["fastapi"],
      "databases": ["postgresql"],
      "infrastructure": ["docker", "nginx"]
    }
  }'

Python SDK

updated = client.projects.update(
    "c1d2e3f4-a5b6-7890-cdef-123456789012",
    stack={"languages": ["python"], "frameworks": ["fastapi"], "databases": ["postgresql"]},
)

---

Delete Project

DELETE /v1/projects/{id}

Soft-delete a project. Memories scoped to this project are preserved but the project is excluded from listings.

Path Parameters

Parameter	Type	Description
id	UUID	Project ID

Response

Returns 204 No Content on success.

curl

curl -X DELETE https://api.memoryrelay.net/v1/projects/c1d2e3f4-a5b6-7890-cdef-123456789012 \
  -H "Authorization: Bearer $MEMORYRELAY_API_KEY"

Python SDK

client.projects.delete("c1d2e3f4-a5b6-7890-cdef-123456789012")

---

Get Project Context

GET /v1/projects/{slug}/context

Retrieve aggregated context for a project, including recent memories, active decisions, and adopted patterns. This is the primary endpoint for building a project-aware prompt for your AI agent.

Path Parameters

Parameter	Type	Description
slug	string	Project slug (e.g. my-api)

Response

{
  "project": {
    "id": "c1d2e3f4-a5b6-7890-cdef-123456789012",
    "slug": "my-api",
    "name": "My API Service",
    "stack": {
      "languages": ["python"],
      "frameworks": ["fastapi"]
    }
  },
  "memories": [
    {
      "id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
      "content": "API uses FastAPI with async SQLAlchemy 2.0",
      "memory_type": "fact",
      "importance": 0.8,
      "created_at": 1710672000
    }
  ],
  "decisions": [
    {
      "id": "f1a2b3c4-d5e6-7890-abcd-ef1234567890",
      "title": "Use PostgreSQL instead of MongoDB",
      "status": "active"
    }
  ],
  "patterns": [
    {
      "id": "b1c2d3e4-f5a6-7890-bcde-f12345678901",
      "title": "Zod validation at API boundaries",
      "category": "validation"
    }
  ]
}

curl

curl https://api.memoryrelay.net/v1/projects/my-api/context \
  -H "Authorization: Bearer $MEMORYRELAY_API_KEY"

Python SDK

context = client.projects.context("my-api")
print(f"Project: {context.project.name}")
print(f"  {len(context.memories)} memories")
print(f"  {len(context.decisions)} active decisions")
print(f"  {len(context.patterns)} adopted patterns")

Agent Bootstrapping

Call GET /v1/projects/{slug}/context at the start of every agent session to load the project's accumulated knowledge. This gives your agent immediate access to past decisions, established patterns, and relevant memories without having to search for them individually.