Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/tboome33/openrouter-fusion-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/mcp/tboome33/openrouter-fusion-mcp/openrouter-fusion)<a href="https://agentmods.dev/mcp/tboome33/openrouter-fusion-mcp/openrouter-fusion"><img src="https://agentmods.dev/badge/mcp/tboome33/openrouter-fusion-mcp/openrouter-fusion/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/mcp/tboome33/openrouter-fusion-mcp/openrouter-fusion"><img src="https://agentmods.dev/badge/mcp/tboome33/openrouter-fusion-mcp/openrouter-fusion.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
openrouter-fusion scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 17 lines — stays where its author put it; the contents beside it link to each section on GitHub.
{
"openrouter-fusion": {
"command": "npx",
"args": [
"-y",
"openrouter-fusion-mcp"
],
"env": {
"OPENROUTER_API_KEY": "${OPENROUTER_API_KEY}",
"OPENROUTER_FUSION_RESEARCH": "{\"name\":\"research\",\"label\":\"Research (deep)\",\"description\":\"Recherche approfondie, panel frontier diversifié (Opus + GPT + Gemini Pro + DeepSeek V4), juge Opus, web poussé. Le « au secours » deep-research (config DRACO 68.3%). NB: la config #1 DRACO (69%) utilise Fable, indisponible sur ce compte (accès Mythos restreint) — un compte avec accès Fable peut l'ajouter au panel.\",\"analysis_models\":[\"~anthropic/claude-opus-latest\",\"openai/gpt-5.5\",\"~google/gemini-pro-latest\",\"deepseek/deepseek-v4-pro\"],\"judge\":\"~anthropic/claude-opus-latest\",\"reasoning_effort\":\"high\",\"temperature\":null,\"max_tool_calls\":8}",
"OPENROUTER_FUSION_RESEARCHECO": "{\"name\":\"research-eco\",\"label\":\"Research (éco)\",\"description\":\"Recherche approfondie rapport qualité/prix — panel éco (Gemini Flash + Kimi + DeepSeek V4 Pro) jugé par Opus : ~1% du top DRACO à ~50% du coût. Juge Opus conservé (un juge éco = échecs « fusion failed »).\",\"analysis_models\":[\"~google/gemini-flash-latest\",\"~moonshotai/kimi-latest\",\"deepseek/deepseek-v4-pro\"],\"judge\":\"~anthropic/claude-opus-latest\",\"reasoning_effort\":\"high\",\"temperature\":null,\"max_tool_calls\":8}",
"OPENROUTER_FUSION_MATHS": "{\"name\":\"maths\",\"label\":\"Mathématiques\",\"description\":\"Raisonnement mathématique rigoureux — panel de raisonneurs divers (GPT + Gemini Pro + DeepSeek V4 + GLM), juge Opus. Les modèles sollicitent peu le web en maths (le cap reste à 8 par sécurité — un cap trop bas affame le juge Opus). NB: o3 retiré (indisponible sur ce compte).\",\"analysis_models\":[\"openai/gpt-5.5\",\"~google/gemini-pro-latest\",\"deepseek/deepseek-v4-pro\",\"z-ai/glm-5.2\"],\"judge\":\"~anthropic/claude-opus-latest\",\"reasoning_effort\":\"xhigh\",\"temperature\":null,What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 17 lines scan A 35aa0d046993
openrouter-fusion is an MCP server published in the GitHub repository tboome33/openrouter-fusion-mcp (0 stars, last pushed 2mo ago), licensed MIT. Its token cost is not measured: an MCP server costs its tool schemas, not its config file. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other mcp servers, from other repositories
mcp
Detect prompt injection, jailbreaks, and code injection in untrusted text before it reaches an LLM. Runs locally from the @safeprompt.dev/mcp npm package. Needs 3 environment variables to run.
ai-guardian
Governed observability + governance for on-endpoint local LLMs (Ollama): model inventory/provenance, a deterministic secrets/PII/jailbreak prompt scanner, allow/deny policy, and route-through guarding — the complement to IGEL AI Armor, with a built-in governance harness (audit, budget, undo, risk tiers). Runs locally…
promptbudget-mcp
Token-budget-aware text handling: count, truncate, and chunk for LLM prompts. Runs locally from the @mukundakatta/promptbudget-mcp npm package.
compactprompt
Easy-to-use prompt & data compression for LLM workflows. Runs locally from the compactprompt Python package.
promptest-mcp
MCP server "promptest-mcp" as configured in cortexresearch/promptest-mcp. Runs locally from the promptest-mcp npm package.
meta-prompt-mcp
An MCP server that acts as a Prompting Oracle — advice from official Prompting Guides. Runs locally from the meta-prompt-mcp Python package.