Ollama
Named by 71 kept items in this workspace.
38 items name it under tools, 17 under affected products and 16 under classification.technologies.
The items that name it 71
Used to run local AI models that ADStrike detects and offers as choices for its AI orchestrator step.
targeted by PoeLLM malware
auto-update path traversal persistence vulnerability (CVE-2026-42249)
AI backend serving infrastructure affected by disclosures in the AI inference gateway category.
local inference server used to run Qwen2.5-Coder and other open-weight models for the benchmark
Local LLM backend that can be routed to via the local auditing proxy
AI model server honeypotted; ships unauthenticated by default.
Local LLM runner mentioned as a tool where random seeds can be set for sampling, though author notes it was 'not always deterministic' in practice
AI provider used for intelligent payload generation, supports local models
run the model locally via hf.co GGUF reference
local runner for the GGUF model
AI tool with a critical unauthenticated vulnerability.
Local LLM runtime supported for running DarkMoon fully locally
Local model runtime used by the DispatchDesk reference example
inference server target subject to deep compute verification by AgentHound
Optional local/on-premises AI provider for Beetle's AI Assistant/AI Actions
local model backend for the opt-in AI layer when no OpenAI-compatible endpoint is configured
Local LLM inference runtime for OpenOSINT Web UI (fully local, no API key)
local LLM provider option for GhidraGPT that does not require an API key or network access
local model host used to run Argo for free with local/open-source models
Local LLM runtime powering the camera-agent (default qwen2.5:1.5b model, or host-run on macOS/Windows for GPU access).
Runs a real local LLM (qwen2.5:3b-instruct) for LLMVault's Live Mode so attacks target an unpredictable model instead of a scripted bot
Local LLM provider used to run AITM analysis 100% locally and free (e.g. llama3.1 model)
local models used for offline summarization and embeddings within brain0
Local model server T3MP3ST can run fully offline against, default for local/keyless operation
Runs local models with tool-calling support for Pentest-AI when using the local-model route (no cloud API key needed)
Local/offline AI engine powering XORCISM's assistants (threat model Q&A, intel briefs, triage, hunt assistant, copilots)
Local LLM backend used natively by OASIS for chat and embedding models, with auto-pull support
Local model runner supported as a provider, enabling zero-API-cost/air-gapped operation
Supported local model provider for Metis
LLM fallback used by Noir to extract endpoints from unsupported frameworks or custom routing when static rules don't apply
Local LLM runtime supported for running CyberStrike entirely offline in air-gapped environments
Local model server exposing an OpenAI-compatible endpoint, supported by Shannon for local/self-hosted models
Fully local LLM provider option for DockSec's AI analysis so scan data never leaves the machine.
Runs a separate local RAG system indexing 561,000+ chunks of personal notes, Sigma rules, curated repos
Unauthenticated local-inference servers exposed to the internet (Cisco Shodan case study)
Local model runner the agent system prompts can be copied into
instances exposed on port 11434, subject to model listing and model weight extraction
Local model used as the local detector/reasoner for identifying sensitive entities before they are surrogated; runs locally so data never leaves the machine
Runs local LLM models as a provider for Rikugan.
Local LLM server used by Sift's --llm-validate feature to validate classifier matches and filter false positives without sending sensitive data off-device
generation parameters, auth, and custom client kwargs support added/fixed in garak's Ollama generator
generation parameters and auth/custom client kwargs support added
本地模型服务，可配置为 DVLAA 的模型后端
Local LLM runtime option (e.g. llama3.1:70b) for running Lonkero's AI agent mode privately
free, offline local AI model provider for BugHunter's hunting engine
Self-hosted inference server; does not enforce message-ordering restrictions by default
In every workspace you can read 121
AI
Security
All technologies in this workspace.