A
licenseA
qualityD
maintenanceTeaching LLMs that Math.random() is so last century
7
11 npm
2
MIT
Model Salt
Server Details
Randomness from outside the model, so one prompt gives different answers that still fit the ask.
If you are the author of this connector, you can claim ownership by verifying the domain or GitHub account it belongs to. Claimed connector authors can inspect health checks, view analytics, and manage their listing.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- modelsalt/model-salt
- GitHub Stars
- 10
- Server Listing
- Model Salt
- ChatGPT
- –
- Claude
- –
TDQS
A4.2/5.0
Scored across 1 tool
Disambiguation5/5
With only one tool there is no possibility of misselection between tools, so the disambiguation concern is trivially satisfied. The tool's own two-phase protocol (request -> explored) is spelled out explicitly, so the agent knows exactly when and how to invoke it.
Naming Consistency4/5
The single name, model_salt, is snake_case and readable, and no competing convention exists, so consistency holds by default. It is a noun phrase rather than a verb_noun action name, which is a minor departure from typical tool naming patterns.
Tool Count3/5
A one-tool server is inherently thin, but this tool has a narrowly defined single mechanism (injecting controlled randomness into creative requests), so the count is defensible rather than mismatched. It lands in the borderline band because there is no companion surface at all.
Completeness3/5
The tool covers its stated loop end to end: an initial call, a silent exploration step, and a second call returning the final decision, with instructions for producing the actual artifact. However, there is no way to request multiple alternatives, reset/iterate on a prior salt, or inspect the outcome without redoing the flow, leaving minor gaps.
Available Tools
1 toolmodel_saltModel SaltARead-onlyInspect
Call this before any creative or open-ended request (an image, a picture prompt, a story, a poem, a name, an idea, a design, a plan), so the result is varied instead of your usual answer and still exactly what was asked. First call: send the user's request word for word in request (and randomness only if the user asked for more or less variety). It returns an exploring step and an id. Do that step silently, without showing it to the user, and call this tool again with the id and your whole reply to the step in explored. The second call returns what was decided. Then make the result itself in the same reply: for a request for an image, generate the image with your image generation tool and show it (a written prompt is not the result); for writing, write it. The result shows every decided answer.
Inputs: id, keep, request, explored, +1 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Fair, reproducible randomness for agents: dice, coins, integers, list picks, shuffles, with seeds.
Fair, reproducible randomness for agents: dice, coins, integers, list picks, shuffles, with seeds.
Fair, reproducible randomness for agents: dice, coins, integers, list picks, shuffles, with seeds.
One prompt. ~100 independent AI models. One report: consensus, disagreements, and minority insights.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables LLMs like Claude to generate genuine random numbers instead of inventing them, covering dice rolls, coin flips, number draws, random selection from lists, and shuffling. It runs as a free-tier Cloudflare Worker using a cryptographically secure generator, and also exposes a REST API, a demo page, and progress-reporting tools for long-running calls.MIT
- AlicenseNot gradedqualityDmaintenanceProvides true random generation tools for AI agents, including coin flip, dice roller, random numbers, and wheel of fortune, enabling decisions and TTRPG sessions.7 npmMIT
- AlicenseNot gradedqualityDmaintenanceUtilizes the Google Gemini API to generate context-specific categories and diverse options for creative brainstorming. It features a random sampling function to help users bypass predictable AI patterns and discover unexpected, innovative ideas.62 npm1MIT
Try in Browser
Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd
Glama MCP Gateway
Add one secure layer between your agents and this server.