← Platform Overview
Feature Pillar
OpenAI and Anthropic Compatible API
Drop-in API endpoints for external tools, IDEs, and MCP servers.
Connect standard IDE extensions, terminal clients, and MCP servers directly to Engra with full support for /v1/chat/completions and /v1/messages.
Chapters:
Watch on YouTube (@EngraAI) →
OpenAI SDK Drop-In Usage
PYTHON
from openai import OpenAI
client = OpenAI(
base_url="https://api.engra.io/v1",
api_key="engra_sk_live_<YOUR_KEY>"
)
response = client.chat.completions.create(
model="engra-frontier",
messages=[{"role": "user", "content": "Analyze repository architecture"}]
)
View Full Searchable Transcript
Engra provides native API compatibility with standard LLM tooling. Point your IDE plugins or internal tools to
https://api.engra.io/v1 to access Engra intelligence, memory, and verification with existing SDKs.
Frequently Asked Questions
Can I use official OpenAI or Anthropic Python/TypeScript SDKs?
Yes. Configure base_url to https://api.engra.io/v1 and provide your Engra API token.
Does Engra support Model Context Protocol (MCP)?
Yes. Engra includes a built-in MCP server (ENGRA_MCP_SERVER=1 engra mcp) exposing repository memory, execution sandboxes, and verification tools.