# llmxray

Local LLM Observatory — See what your AI is actually doing, token by token.

- **Type:** MCP server
- **Trust:** 60/100 (B), scored on the package rubric
- **Verification:** community-indexed — nobody has claimed this listing
- **Version:** 0.8.11
- **Author:** io.github.LogneBudo
- **License:** Apache-2.0
- **npm:** llmxray
- **Source:** https://github.com/LogneBudo/llmxray
- **Compatible clients:** claude-code, cursor, copilot, gemini (basis: transport)

## Trust

60/100 (B), scored on the package rubric
- Publisher verified: no
- Install scripts: nothing suspicious found
- Prompt-injection scan: not run
- Obfuscation scan: not run
- Evidence age: 0 days

## Security scan

- **Status:** clean
- **Scanned:** 2026-10-21T07:32:26.512Z
- **Version scanned:** 0.8.11
- **CVEs:** none found by OSV at scan time
**Findings**
- injection-shaped content (note) in the `system` tool: References the system prompt

## Tools

40 declared. Statically extracted from the shipped source — a floor on the surface, not a census.
- `help` — Show all available commands
- `clear` — Clear current conversation
- `new` — Start a new chat session
- `summarize` — Ask the model to summarize the conversation
- `save` — Export conversation as JSON or text
- `copy` — Copy last assistant response to clipboard
- `history` — Show recent conversations
- `model` — Switch model or list available models
- `system` — Set the system prompt
- `temperature` — Set temperature (0-2)
- `top_p` — Set top-p / nucleus sampling (0-1)
- `top_k` — Set top-k sampling (1-100)
- `context` — Set context window size
- `repeat_penalty` — Set repeat penalty (1.0-2.0)
- `seed` — Set seed for reproducibility (-1 for random)
- `stop` — Set stop sequences (comma-separated)
- `mirostat` — Set mirostat mode (0=off, 1=v1, 2=v2)
- `format` — Toggle JSON output format
- `reset` — Reset all settings to defaults
- `rag` — Toggle RAG context on/off
- `tools` — Open tools settings panel
- `status` — Show Ollama connection status and model info
- `tokens` — Show token count for current session
- `speed` — Show average generation speed across sessions
- `compare` — Navigate to the Compare page
- `session` — View current session details
- `remember` — Remember a fact about you across conversations
- `forget` — Forget a previously remembered fact
- `memories` — Show all remembered facts
- `nomic-embed-text` — Nomic — 137M params, fast & good
- `mxbai-embed-large` — mixedbread.ai — 335M params, high quality
- `all-minilm` — MiniLM — 23M params, very lightweight
- `snowflake-arctic-embed` — Snowflake — 110M params, balanced
- `llama3.2` — Meta Llama 3.2 -- fast general-purpose chat
- `llama3.1` — Meta Llama 3.1 -- strong reasoning and instruction following
- `gemma3` — Google Gemma 3 -- efficient and multilingual
- `gemma2` — Google Gemma 2 -- balanced performance
- `mistral` — Mistral 7B -- strong European model
- `phi4` — Microsoft Phi-4 -- small but capable
- `qwen2.5` — Alibaba Qwen 2.5 -- multilingual powerhouse

## Install

**Verdict: review** — Installable, but 1 thing to check first: No publisher has proved control of this listing; it is indexed, not vouched for.
**Cautions** (coverage gaps and advisories — never blocking)
- No publisher has proved control of this listing; it is indexed, not vouched for.
**Config** (claude-code):
```json
"{\n  \"mcpServers\": {\n    \"llmxray\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"-y\",\n        \"llmxray\"\n      ]\n    }\n  }\n}"
```

## Blast radius

Contained to moderate — no credential declaration found, from the publisher, the upstream registry, or the README. Known so far: runs on your machine; read-only tool surface.
- Floor 13, ceiling 31 (tier: unknown)
- `unknown` means the floor and ceiling land in different bands — not measured enough to name one. It does not mean low.
- This is impact, not likelihood. A high radius is not a defect: a filesystem server is supposed to write files. It is never part of the trust score.

## Machine-readable views of this entry

- Signed JSON: https://forgeregistry.com/api/v1/packages/llmxray
- Install plan: https://forgeregistry.com/api/v1/packages/llmxray/install-plan
- Alternatives: https://forgeregistry.com/api/v1/alternatives/llmxray
- HTML page: https://forgeregistry.com/registry/llmxray
- MCP: POST https://forgeregistry.com/api/mcp → `forge_get_package` / `forge_install_plan`

## About this document

Generated by Forge (https://forgeregistry.com) — a compact rendering of the same record served, signed, at the JSON URL above. Trust and scan facts are the registry's own measurements; anything Forge did not measure is named as unmeasured rather than omitted.