Skip to content

dependencies-and-llm-providers.md

Latest commit

 

History

History
121 lines (92 loc) · 4.98 KB

dependencies-and-llm-providers.md

File metadata and controls

121 lines (92 loc) · 4.98 KB
Raw

Dependencies and LLM Providers

Core Runtime Dependencies

All versions from pyproject.toml. Requires Python >= 3.10.

Web Framework and API

Package Version Purpose
fastapi >= 0.125.0 REST API framework
uvicorn[standard] >= 0.29.0 ASGI server
pydantic >= 2.6.0 Data validation and serialization
python-multipart >= 0.0.6 File upload handling
httpx >= 0.28.1 Async HTTP client

CLI and TUI

Package Version Purpose
click >= 8.1.0 CLI framework
rich >= 13.0.0 Terminal formatting
textual >= 1.0.0 Interactive TUI (policy configurator)
tabulate >= 0.9.0 Table output formatting

Analysis and Detection

Package Version Purpose
yara-x >= 1.12.0 Pattern-matching rule engine
magika >= 0.6.0 AI-powered file type detection (200+ types)
pdfid >= 1.1.0 Structural PDF analysis (JS, OpenAction, Launch)
oletools >= 0.60.1 Office document macro/VBA detection
confusable-homoglyphs >= 3.3.0 Unicode homoglyph attack detection

Data and Configuration

Package Version Purpose
PyYAML >= 6.0.1 YAML parsing (policies, rules)
python-frontmatter >= 1.0.0 SKILL.md frontmatter parsing
python-dotenv >= 1.0.0 Environment variable loading from .env

LLM SDKs

Package Version Purpose
anthropic >= 0.40.0 Anthropic Claude SDK
openai >= 1.0.0 OpenAI SDK
litellm >= 1.77.0 Multi-provider LLM routing
google-genai optional via [google] Google AI Studio / Gemini SDK
google-generativeai optional via [google] Legacy Google Generative AI SDK compatibility

Optional Provider Extras

Install only what you need:

# AWS Bedrock
pip install "cisco-ai-skill-scanner[bedrock]"

# Google AI Studio / Gemini
pip install "cisco-ai-skill-scanner[google]"

# Google Vertex AI
pip install "cisco-ai-skill-scanner[vertex]"

# Azure OpenAI
pip install "cisco-ai-skill-scanner[azure]"

# All provider extras
pip install "cisco-ai-skill-scanner[all]"
Extra Package Version Purpose
google google-genai, google-generativeai varies Google AI Studio / Gemini SDK support
bedrock boto3 >= 1.28.57 AWS Bedrock IAM credential support
vertex google-cloud-aiplatform >= 1.38.0 Google Vertex AI support
azure azure-identity >= 1.15.0 Azure managed identity auth
all all of the above Install all provider extras

Supported LLM Providers

Model Naming

Set SKILL_SCANNER_LLM_MODEL using the provider prefix convention:

Provider Model example Notes
Anthropic anthropic/claude-sonnet-4-20250514 Default provider
OpenAI openai/gpt-4o
OpenAI-compatible custom endpoint Cloud-Gemini-3.1-Pro with SKILL_SCANNER_LLM_PROVIDER=openai Uses SKILL_SCANNER_LLM_BASE_URL
AWS Bedrock bedrock/anthropic.claude-sonnet-4-20250514-v1:0 Requires [bedrock] extra or API key
Google Vertex AI vertex_ai/gemini-2.5-pro Requires [vertex] extra
Google AI Studio gemini/gemini-2.5-flash Requires [google] extra
Azure OpenAI azure/my-deployment-name Requires [azure] extra
Ollama (local) ollama/llama3 No API key needed

For OpenAI and OpenAI-compatible custom endpoints, SKILL_SCANNER_LLM_USER can set the raw Chat Completions user request field. The value is passed through unchanged and is ignored for non-OpenAI routes.

Authentication

Provider Auth method Required env vars
Anthropic API key SKILL_SCANNER_LLM_API_KEY
OpenAI API key SKILL_SCANNER_LLM_API_KEY
OpenAI-compatible custom endpoint API key + endpoint SKILL_SCANNER_LLM_API_KEY, SKILL_SCANNER_LLM_PROVIDER=openai, SKILL_SCANNER_LLM_BASE_URL
AWS Bedrock (API key) API key SKILL_SCANNER_LLM_API_KEY
AWS Bedrock (IAM) AWS credentials AWS_REGION, AWS_PROFILE (optional: AWS_SESSION_TOKEN)
Google Vertex AI (service account) Service account key file GOOGLE_APPLICATION_CREDENTIALS
Google Vertex AI (ambient ADC) Workload Identity / attached service account none -- falls back automatically, like Bedrock IAM
Google AI Studio API key SKILL_SCANNER_LLM_API_KEY (auto-sets GEMINI_API_KEY)
Azure OpenAI API key + endpoint SKILL_SCANNER_LLM_API_KEY, SKILL_SCANNER_LLM_BASE_URL, SKILL_SCANNER_LLM_API_VERSION
Ollama None —

Related