Search and discover AI models available in the Runware platform. This tool enables discovery of available models on the Runware platform, providing powerful search and filtering capabilities. Whether exploring public models from the community or managing private models within your organization, t...
Risk signalsHigh parameter count (10 properties)
Part of the Mcp Runware server.
Free to start. No card required.
AI agents call modelSearch to retrieve information from Mcp Runware without modifying any data. This is common in research, monitoring, and reporting workflows where the agent needs context before taking action. Because read operations don't change state, they are generally safe to allow without restrictions -- but you may still want rate limits to control API costs.
Even though modelSearch only reads data, uncontrolled read access can leak sensitive information or rack up API costs. An agent caught in a retry loop could make thousands of calls per minute. A rate limit gives you a safety net without blocking legitimate use.
Read-only tools are safe to allow by default. No rate limit needed unless you want to control costs.
{
"version": "1",
"default": "deny",
"tools": {
"modelSearch": {}
}
}See the full Mcp Runware policy for all 11 tools.
These attack patterns abuse exactly the kind of access modelSearch gives an agent. Each links to the full case and the policy that stops it:
Other read tools across the catalogue. The same approach applies to each: allow, with a rate cap to control cost.
Search and discover AI models available in the Runware platform. This tool enables discovery of available models on the Runware platform, providing powerful search and filtering capabilities. Whether exploring public models from the community or managing private models within your organization, this API helps find the perfect model for any image generation task. Models discovered through this tool can be immediately used in image generation tasks by referencing their AIR identifiers. This enables dynamic model selection in applications and helps discover new models for specific artistic styles. Args: search (str, optional): Search term to filter models. The search is performed across multiple fields: - Model name as exact phrase (boost: 10) - Model AIR identifier with wildcard matching (boost: 5) - Model name with wildcard matching (boost: 5) - Model version (exact word matching) - Model tags (exact word matching) The search is case-insensitive and results are ordered by relevance. tags (List[str], optional): Filter models by matching any of the provided tags. Models that contain at least one of these tags will be included in the results. category (str, optional): Filter models by their category: - "checkpoint": Base models that serve as the foundation for image generation - "lora": LoRA (Low-Rank Adaptation) models for specific styles or concepts - "lycoris": Alternative to LoRA models with different adaptation techniques - "controlnet": Models for guided image generation with specific conditions - "vae": Variational Autoencoders for improving image quality and details - "embeddings": Textual embeddings for adding new concepts to the model's vocabulary type (str, optional): Filter checkpoint models by their type (only applicable when category is "checkpoint"): - "base": Standard models for general image generation - "inpainting": Models for filling in or modifying parts of existing images - "refiner": Models that improve the quality and details of generated images architecture (str, optional): Filter models by their architecture: FLUX Models: "flux1s", "flux1d", "fluxpro", "fluxultra", "fluxkontextdev", "fluxkontextpro", "fluxkontextmax" Imagen Models: "imagen3", "imagen3fast", "imagen4preview", "imagen4ultra", "imagen4fast" HiDream Models: "hidreamfast", "hidreamdev", "hidreamfull" SD Models: "sd1x", "sdhyper", "sd1xlcm", "sdxl", "sdxllcm", "sdxldistilled", "sdxlhyper", "sdxllightning", "sdxlturbo", "sd3" Other: "pony" conditioning (str, optional): Filter ControlNet models by their conditioning type (only applicable when category is "controlnet"): Edge Detection: "blur", "canny", "hed", "lineart", "softedge" Spatial: "depth", "normal", "seg" Creative: "inpaint", "inpaintdepth", "pix2pix", "scribble", "sketch" Specialized: "gray", "lowquality", "openmlsd", "openpose", "outfit", "qrcode", "shuffle", "tile" visibility (str): Filter models by visibility status and ownership: - "public": Show only your organization's public models - "private": Show only your organization's private models - "all": Show both community models and all your organization's models (default) limit (int): Maximum number of items to return (1-100, default: 20). Used for pagination. offset (int): Number of items to skip in the result set (min: 0, default: 0). Used for pagination. taskUUID (UUID, optional): Unique task identifier Returns: dict: A dictionary containing the model search results with status, message, result data, and comprehensive model information including AIR identifiers for immediate use. Note: For optimal search performance, consider using specific filters to narrow results and combining multiple criteria to find the most relevant models. Results are returned in paginated format for efficient processing of large result sets.. It is categorised as a Read tool in the Mcp Runware MCP Server, which means it retrieves data without modifying state.
Register the Mcp Runware MCP server in PolicyLayer and add a rule for modelSearch: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Mcp Runware. Nothing to install.
modelSearch is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the modelSearch rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for modelSearch. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
modelSearch is provided by the Mcp Runware MCP server (elijahdev0/mcp-runware). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 11 Mcp Runware tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.