Local LLM Observatory — See what your AI is actually doing, token by token.
Inferred from the transports this listing declares (stdio). A client not listed here hasn’t been ruled out — it just isn’t something Forge can confirm.
Verification confirms publisher identity (repo ownership), not code safety. The security scan covers known CVEs and suspicious install scripts.
Read out of the source npm actually ships, at scan time. The package was never executed. Tools registered dynamically at runtime, or hidden inside bundled or minified code, can be missed — so this is a floor on the tool surface, not a complete census of it.
helpShow all available commandsShow all available commands
No input schema was published for this tool.
clearClear current conversationClear current conversation
No input schema was published for this tool.
newStart a new chat sessionStart a new chat session
No input schema was published for this tool.
summarizeAsk the model to summarize the conversationAsk the model to summarize the conversation
No input schema was published for this tool.
saveExport conversation as JSON or textExport conversation as JSON or text
No input schema was published for this tool.
copyCopy last assistant response to clipboardCopy last assistant response to clipboard
No input schema was published for this tool.
historyShow recent conversationsShow recent conversations
No input schema was published for this tool.
modelSwitch model or list available modelsSwitch model or list available models
No input schema was published for this tool.
systemSet the system promptSet the system prompt
Set the system promptNo input schema was published for this tool.
temperatureSet temperature (0-2)Set temperature (0-2)
No input schema was published for this tool.
top_pSet top-p / nucleus sampling (0-1)Set top-p / nucleus sampling (0-1)
No input schema was published for this tool.
top_kSet top-k sampling (1-100)Set top-k sampling (1-100)
No input schema was published for this tool.
contextSet context window sizeSet context window size
No input schema was published for this tool.
repeat_penaltySet repeat penalty (1.0-2.0)Set repeat penalty (1.0-2.0)
No input schema was published for this tool.
seedSet seed for reproducibility (-1 for random)Set seed for reproducibility (-1 for random)
No input schema was published for this tool.
stopSet stop sequences (comma-separated)Set stop sequences (comma-separated)
No input schema was published for this tool.
mirostatSet mirostat mode (0=off, 1=v1, 2=v2)Set mirostat mode (0=off, 1=v1, 2=v2)
No input schema was published for this tool.
formatToggle JSON output formatToggle JSON output format
No input schema was published for this tool.
resetReset all settings to defaultsReset all settings to defaults
No input schema was published for this tool.
ragToggle RAG context on/offToggle RAG context on/off
No input schema was published for this tool.
toolsOpen tools settings panelOpen tools settings panel
No input schema was published for this tool.
statusShow Ollama connection status and model infoShow Ollama connection status and model info
No input schema was published for this tool.
tokensShow token count for current sessionShow token count for current session
No input schema was published for this tool.
speedShow average generation speed across sessionsShow average generation speed across sessions
No input schema was published for this tool.
compareNavigate to the Compare pageNavigate to the Compare page
No input schema was published for this tool.
sessionView current session detailsView current session details
No input schema was published for this tool.
rememberRemember a fact about you across conversationsRemember a fact about you across conversations
No input schema was published for this tool.
forgetForget a previously remembered factForget a previously remembered fact
No input schema was published for this tool.
memoriesShow all remembered factsShow all remembered facts
No input schema was published for this tool.
nomic-embed-textNomic — 137M params, fast & goodNomic — 137M params, fast & good
No input schema was published for this tool.
mxbai-embed-largemixedbread.ai — 335M params, high qualitymixedbread.ai — 335M params, high quality
No input schema was published for this tool.
all-minilmMiniLM — 23M params, very lightweightMiniLM — 23M params, very lightweight
No input schema was published for this tool.
snowflake-arctic-embedSnowflake — 110M params, balancedSnowflake — 110M params, balanced
No input schema was published for this tool.
llama3.2Meta Llama 3.2 -- fast general-purpose chatMeta Llama 3.2 -- fast general-purpose chat
No input schema was published for this tool.
llama3.1Meta Llama 3.1 -- strong reasoning and instruction followingMeta Llama 3.1 -- strong reasoning and instruction following
No input schema was published for this tool.
gemma3Google Gemma 3 -- efficient and multilingualGoogle Gemma 3 -- efficient and multilingual
No input schema was published for this tool.
gemma2Google Gemma 2 -- balanced performanceGoogle Gemma 2 -- balanced performance
No input schema was published for this tool.
mistralMistral 7B -- strong European modelMistral 7B -- strong European model
No input schema was published for this tool.
phi4Microsoft Phi-4 -- small but capableMicrosoft Phi-4 -- small but capable
No input schema was published for this tool.
qwen2.5Alibaba Qwen 2.5 -- multilingual powerhouseAlibaba Qwen 2.5 -- multilingual powerhouse
No input schema was published for this tool.
40 of 40 tools published a description.
Tool names and descriptions are written by the publisher and shown verbatim as inert text. They are the strings an MCP client passes to a model, so Forge scans them for prompt-injection patterns — any finding appears with the security scan above. “Privileged” is a keyword match on the tool name, not an audit of what the tool does: a benign-sounding name can still do anything.
Local LLM Observatory — See what your AI is actually doing, token by token.
Linked names open Forge’s index of every entry observed exposing that tool. Browse all indexed tools.
The crawl stopped at the 60-package limit. The rest of the tree was never resolved.
36 more resolved packages are not drawn here (display cap: 24). Every dependency carrying an advisory is drawn regardless of the cap. Full inventory (CycloneDX SBOM)
38 declared dependencies never landed in the tree. They are missing from Forge's resolution, not from the package.
+26 more not listed. The counts by reason above cover all of them.
Not followed: peerDependencies. This tree covers runtime dependencies only, so anything those pull in was never resolved.