Security audit
Security checks for vulnerabilities and agentic risk
This is a legitimate shared-memory setup, but it broadly indexes private agent files into a persistent cross-agent graph with weak scoping and review controls.
Install only if you intentionally want a shared, persistent memory service across agents. Before using it, pin the Docker image, protect API keys, disable cron until reviewed, use explicit file allowlists, redact secrets and personal data, decide which agents may read and write, and document how to remove mistakenly indexed data.
KEY = "your-secure-api-key"
def insert(text, description="manual insert"):
r = requests.post(f"{API}/documents/text",
headers={"Authorization": f"Bearer {KEY}", "Content-Type": "application/json"},
json={"text": text, "description": description},
timeout=120)# LLM (for graph construction and queries) LLM_BINDING=openai LLM_MODEL=gpt-4.1-nano LLM_BINDING_HOST=https://api.openai.com/v1 LLM_BINDING_API_KEY=sk-your-api-key # Embeddings (for vector search)
EMBEDDING_BINDING=openai EMBEDDING_MODEL=text-embedding-3-small EMBEDDING_DIM=1536 EMBEDDING_BINDING_HOST=https://api.openai.com/v1 EMBEDDING_BINDING_API_KEY=sk-your-api-key # Performance
## Step 6: Auto-Index New Daily Logs (Optional) Create a cron job to index new daily logs automatically: ```bash #!/bin/bash
No suspicious patterns detected.