Skip to content

    Type a page, a command such as keryn scan, or a rule id.

    inventory-aibom-v2

    View as Markdown

    95 rules · Keryn Software Licence · rules 0.8.0 · all packs

    How to read a rule
    Severity
    How bad the finding is if it is real: error fails a default gate, warning and info do not.
    Confidence
    P0 exploitable as written; P1 dangerous in context; P2 a heuristic or a posture check: confirm by reading the code.
    False-positive risk
    low: an exact API or flag, so hits are almost always real; medium: the pattern is legitimate in many programs, so check each hit; high: it finds a lead to confirm, so expect benign hits.
    Edge
    The relationship the finding sits on. reaches: content or traffic gets to a thing; calls: one thing invokes another; reads: one thing reads from another; writes: one thing writes to another; sends: data leaves for another party; grants: authority is given or left unchecked; depends_on: something relies on a component it does not verify; trusts: one party believes another without evidence. The boundary it crosses, when there is one: user->prompt (untrusted input reaching a prompt); model->code (model output reaching code, a shell, a query or a page); secret->outside (a secret leaving the trusted side); supply-chain (code, models or prompts pulled in from outside).
    Taxonomy chips
    References for readers who work in OWASP LLM 2025, OWASP Agentic, CWE or NIST AI RMF. They say where a rule belongs, not that a framework is satisfied.

    info generic azure

    References the Azure OpenAI API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic azure

    References the Azure AI Foundry model inference API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://learn.microsoft.com/en-us/azure/foundry/foundry-models/concepts/endpoints

    info generic aws

    References the Amazon Bedrock runtime API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://raw.githubusercontent.com/boto/botocore/develop/botocore/data/bedrock-runtime/2023-09-30/service-2.json

    info generic aws

    References the Amazon Bedrock Agents runtime API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic google

    References the Google Vertex AI API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://docs.cloud.google.com/gemini-enterprise-agent-platform/reference/rest

    info generic cohere

    References the Cohere (legacy api.cohere.ai host) API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic mistral

    References the Mistral Codestral API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic deepseek

    References the DeepSeek API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic fireworks

    References the Fireworks AI API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic perplexity

    References the Perplexity API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic together

    References the Together AI (api.together.ai host) API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic xai

    References the xAI API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic replicate

    References the Replicate API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic huggingface

    References the Hugging Face inference API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic voyage

    References the Voyage AI API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic jina

    References the Jina AI API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic cerebras

    References the Cerebras Inference API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic sambanova

    References the SambaNova Cloud API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic nvidia

    References the NVIDIA NIM hosted API API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic alibaba

    References the Alibaba Cloud DashScope / Model Studio API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic moonshot

    References the Moonshot AI (Kimi) API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic minimax

    References the MiniMax API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic zhipu

    References the Zhipu AI / Z.ai API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://docs.z.ai/guides/overview/quick-start

    info generic baidu

    References the Baidu Qianfan API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic elevenlabs

    References the ElevenLabs API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic deepgram

    References the Deepgram API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic assemblyai

    References the AssemblyAI API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://www.npmjs.com/package/assemblyai

    info generic stability

    References the Stability AI API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic fal

    References the fal.ai model endpoints API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/fal-client/

    info generic runway

    References the Runway API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/runwayml/

    info generic tavily

    References the Tavily search API API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic exa

    References the Exa search API API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://www.npmjs.com/package/exa-js

    info generic serper

    References the Serper search API API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic langfuse

    References the Langfuse Cloud API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic langchain

    References the LangSmith Cloud API host.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://www.npmjs.com/package/langsmith

    info generic cohere

    References cohere.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic mistral

    References mistral.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic groq

    References groq.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic together

    References together.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic fireworks

    References fireworks.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic replicate

    References replicate.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic voyage

    References voyageai.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic openrouter

    References openrouter.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic azure

    References azure-ai-inference.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic aws-bedrock

    References bedrock-runtime.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://raw.githubusercontent.com/openai/openai-python/main/src/openai/lib/azure.py · https://raw.githubusercontent.com/boto/botocore/develop/botocore/data/bedrock-runtime/2023-09-30/service-2.json

    info generic google-vertex

    References vertex-ai.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://raw.githubusercontent.com/openai/openai-python/main/src/openai/lib/azure.py

    info generic dspy

    References dspy.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic haystack

    References haystack.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic semantic-kernel

    References semantic-kernel.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic smolagents

    References smolagents.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic pydantic-ai

    References pydantic-ai.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic openai-agents

    References openai-agents.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic inventory-aibom

    References structured-output-library.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic inventory-aibom

    References llm-observability.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic sentence-transformers

    References sentence-transformers.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic inventory-aibom

    References local-inference-engine.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic inventory-aibom

    References ml-framework.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic vercel-ai-sdk

    References vercel-ai-sdk-provider.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic genkit

    References genkit.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic copilotkit

    References copilotkit.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic go

    References a Go AI SDK or agent module.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic java, kotlin

    References a Java/Kotlin AI SDK or framework.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic dotnet

    References a .NET AI SDK or framework.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic php

    References a PHP AI client library.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://raw.githubusercontent.com/theodo-group/LLPhant/main/composer.json

    info generic ruby

    References a Ruby AI gem.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic rust

    References a Rust AI crate.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic swift

    References a Swift AI SDK or on-device ML framework.

    Confidence P2: a heuristic or a posture check: confirm by reading the code.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic dart, flutter

    References a Dart/Flutter AI package.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic cpp

    Includes a C/C++ model-runtime header (llama.cpp, ggml, ONNX Runtime, liboai).

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic inventory-aibom

    References an LLM tokenizer library (tiktoken, tokenizers, gpt-tokenizer).

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic inventory-aibom

    Hard-coded open embedding model identifier.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic gradio

    References gradio.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic bentoml

    References bentoml.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic mlflow

    References mlflow.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic ultralytics

    References ultralytics.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic spacy

    References spacy.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic onnx

    References onnx.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic dvc

    References dvc.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic ray

    References ray.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    info generic claude-agent-sdk

    References claude-agent-sdk.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/claude-agent-sdk/

    info generic google-adk

    References google-adk.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/google-adk/

    info generic strands-agents

    References strands-agents.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/strands-agents/

    info generic bedrock-agentcore

    References bedrock-agentcore.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/bedrock-agentcore/

    info generic microsoft-agent-framework

    References microsoft-agent-framework.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/agent-framework-core/

    info generic agno

    References agno.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/agno/

    info generic mem0

    References mem0.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/mem0ai/

    info generic zep

    References zep.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/zep-cloud/

    info generic letta

    References letta.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/letta-client/

    info generic e2b

    References e2b.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/e2b/

    info generic daytona

    References daytona.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/daytona/

    info generic modal

    References modal.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/modal/

    info generic browser-use

    References browser-use.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/browser-use/

    info generic a2a

    References the A2A (Agent2Agent) protocol SDK.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/a2a-sdk/

    info generic ag-ui

    References the AG-UI (Agent-User Interaction) protocol SDK.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/ag-ui-protocol/

    info generic unstructured

    References unstructured.

    Confidence P1: dangerous in context.

    Edge inspected: depends_on something relies on a component it does not verify; it crosses the supply-chain boundary (code, models or prompts pulled in from outside).

    References: https://pypi.org/project/unstructured/