Quickstart

Connect your first project and load context into an agent in less than 5 minutes.

1. Create the project

From the web UI at /projects/create, enter:

Name and Code (e.g. my-project) - the only required fields.
Repository URL (GitHub, GitLab or Gitea) - optional.
Git token if the repo is private - optional.
Notion URL + integration token - optional.

💡 You can create an empty project without external sources. In that case, it starts empty and you can register memories manually via API or the dashboard.

After saving, SoluCortex shows two values you need to store immediately:

Project ID - the project UUID (always visible in the dashboard).
API key (scx_...) - copy it now, it is only shown once.

⚠️ The API key cannot be recovered later. If you lose it, generate a new one from the project dashboard. The Project ID is always visible in the URL and on the project page.

2. Save the credentials

In your project, create or edit ai/.env.ai:

bash

SOLUCORTEX_URL=https://solucortex.ai
SOLUCORTEX_PROJECT_ID=<uuid-of-the-project>
SOLUCORTEX_API_KEY=scx_<your-api-key>

🔒 ai/.env.ai must be in .gitignore. Never commit credentials.

3. Initial import (if you connected sources)

If you configured Git or Notion, the worker imports memories in the background (~1–3 min). Check the dashboard for memories with pending status and approve them from /memories → Approve all in filter.

If you created an empty project, skip this step - the project is already ready to receive memories.

4. Load context into your agent

bash

source ai/.env.ai

curl -s -X POST "$SOLUCORTEX_URL/api/v1/context/build" \
  -H "Authorization: Bearer $SOLUCORTEX_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json" \
  -d "{
    \"project_id\": \"$SOLUCORTEX_PROJECT_ID\",
    \"query\": \"implement authentication with API keys\"
  }"

The response returns the most relevant memories, ready to paste into the agent context.

5. Register your first memory

bash

curl -s -X POST "$SOLUCORTEX_URL/api/v1/memories" \
  -H "Authorization: Bearer $SOLUCORTEX_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json" \
  -d "{
    \"project_id\": \"$SOLUCORTEX_PROJECT_ID\",
    \"type\": \"decision\",
    \"title\": \"Use pgvector for semantic search\",
    \"content\": \"pgvector was chosen over Pinecone to keep everything in PostgreSQL and avoid external dependencies.\",
    \"importance\": 8
  }"

✓ Done! You now have your first connected project working. The memory was registered and will be available in future semantic searches once its embedding is generated.