Aegis Research SDK

Python SDK for the Aegis Research API - AI-powered web research as a service.

Installation

pip install aegis-research

Quick Start

from aegis_research import AegisResearch

# Initialize client
client = AegisResearch(api_key="res_your_api_key")

# Execute research
result = client.research("Best practices for API rate limiting in 2025")

print(result.summary)
for finding in result.key_findings:
    print(f"- {finding}")

Features

• AI-Powered Research: Get synthesized research results, not just search links
• Multiple Depth Levels: Choose shallow (1 credit), medium (3 credits), or deep (10 credits) research
• Caching: Results cached for 24 hours to save credits
• Source Citations: Every claim backed by sources

Usage

Research

# Basic research
result = client.research("What is quantum computing?")

# Deep research with more sources
result = client.research(
    "Impact of AI on healthcare in 2025",
    depth="deep",  # 10 credits, 10+ sources
    use_cache=False  # Force fresh research
)

# Include specific URLs
result = client.research(
    "Compare FastAPI vs Flask",
    urls=["https://fastapi.tiangolo.com", "https://flask.palletsprojects.com"]
)

Check Credits

status = client.credits()
print(f"Credits remaining: {status.credits_remaining}")
print(f"Tier: {status.tier}")

Research History

history = client.history(limit=10)
for item in history:
    print(f"{item['id']}: {item['topic']}")

Get Previous Result

result = client.get_research("res_abc123")

Response Format

ResearchResult:
    id: str                     # "res_abc123"
    topic: str                  # Original query
    status: str                 # "completed", "cached", "failed"
    summary: str                # Executive summary (2-3 sentences)
    key_findings: List[str]     # Main findings as bullet points
    detailed_analysis: str      # Full analysis
    sources: List[Source]       # Cited sources
    source_count: int           # Number of sources
    depth: str                  # "shallow", "medium", "deep"
    cached: bool                # Whether result was cached
    credits_used: int           # Credits consumed
    duration_ms: int            # Time taken

Depth Levels

Real-World Examples

The examples/ directory contains ready-to-use scripts for common use cases:

# Set your API key
export AEGIS_API_KEY=res_your_key

# Run an example
python examples/competitor_analysis.py

Error Handling

from aegis_research import (
    AegisError,
    AuthenticationError,
    RateLimitError,
    InsufficientCreditsError,
)

try:
    result = client.research("topic")
except AuthenticationError:
    print("Invalid API key")
except RateLimitError as e:
    print(f"Rate limited: {e}")
except InsufficientCreditsError:
    print("Not enough credits")
except AegisError as e:
    print(f"API error: {e}")

Pricing

Get your API key at aegisagent.ai/research

License

MIT