Skip to main content
Glama
collapseindex

CI-1T Prediction Stability Engine

onboarding

Get started with the CI-1T Prediction Stability Engine. Returns setup instructions, available tools, pricing, and links for new users.

Instructions

Welcome guide for new users. Returns setup instructions, available tools, pricing, and links. No auth required. Call this when someone is new to CI-1T or asks how to get started.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Implementation Reference

  • This is the tool handler for the 'onboarding' tool. It's a local (no auth) MCP tool that returns a comprehensive welcome guide for new CI-1T users, including status, setup instructions, environment variables, config examples, available tools list, pricing, and links.
    server.tool(
      "onboarding",
      "Welcome guide for new users. Returns setup instructions, available tools, pricing, and links. No auth required. Call this when someone is new to CI-1T or asks how to get started.",
      {},
      async () => {
        const hasApiKey = !!API_KEY;
        const ready = hasApiKey;
    
        return {
          content: [
            {
              type: "text" as const,
              text: JSON.stringify(
                {
                  welcome: "Welcome to CI-1T — the prediction stability engine.",
                  status: {
                    api_key: hasApiKey ? "✓ configured" : "✗ not set",
                    ready: hasApiKey,
                  },
                  what_is_ci1t:
                    "CI-1T measures whether ML model predictions are stable or collapsing. It computes the Collapse Index (CI) — a Q0.16 fixed-point metric where 0 = perfectly stable and 65535 = total collapse. Useful for detecting silent failures, ghost predictions (confident but wrong), and model drift.",
                  getting_started: [
                    `1. Create a free account at ${SIGNUP_URL} — you get 1,000 free credits on signup`,
                    "2. Go to Dashboard → API Keys → Create a new key (starts with ci_...)",
                    "3. Set CI1T_API_KEY in your MCP client config — that's the only credential you need",
                    "4. Start evaluating — your free credits cover 1,000 episodes",
                  ],
                  environment_variables: {
                    CI1T_API_KEY: {
                      required_for: "All authenticated tools",
                      format: "ci_... (32 chars)",
                      description: "Your API key — single credential for evaluations, fleet sessions, API key management, and billing",
                    },
                    CI1T_BASE_URL: {
                      required_for: "optional",
                      format: "URL",
                      description: `API base URL (default: ${BASE_URL})`,
                    },
                  },
                  config_example: {
                    claude_desktop: {
                      mcpServers: {
                        ci1t: {
                          command: "npx",
                          args: ["-y", "@collapseindex/ci1t-mcp"],
                          env: {
                            CI1T_API_KEY: "ci_your_key_here",
                          },
                        },
                      },
                    },
                  },
                  tools_available_now: [
                    "onboarding — this guide (no auth)",
                    "interpret_scores — statistical breakdown of scores (no auth)",
                    "convert_scores — float ↔ Q0.16 conversion (no auth)",
                    "generate_config — integration boilerplate (no auth)",
                    ...(hasApiKey
                      ? [
                          "evaluate — run stability analysis (API key ✓)",
                          "fleet_evaluate — multi-node fleet analysis (API key ✓)",
                          "probe — test LLM stability (API key ✓)",
                          "health — engine status (API key ✓)",
                          "fleet_session_* — persistent sessions (API key ✓)",
                          "API key management (API key ✓)",
                          "get_invoices — billing (API key ✓)",
                        ]
                      : []),
                  ],
                  pricing: {
                    free_credits: "1,000 credits on signup — no credit card required",
                    evaluate: "1 credit per episode",
                    fleet_evaluate: "1 credit per episode per node",
                    probe: "1 credit per probe",
                    utility_tools: "Free — no credits, no auth",
                  },
                  links: {
                    dashboard: SIGNUP_URL,
                    docs: DOCS_URL,
                    github: "https://github.com/collapseindex",
                  },
                },
                null,
                2
              ),
            },
          ],
        };
      }
    );
  • The schema is defined as an empty object ({}) — no input parameters are required for the onboarding tool. It takes no arguments from the user.
    server.tool(
      "onboarding",
      "Welcome guide for new users. Returns setup instructions, available tools, pricing, and links. No auth required. Call this when someone is new to CI-1T or asks how to get started.",
      {},
  • src/index.ts:369-458 (registration)
    The onboarding tool is registered via server.tool() on the MCP server instance, with the name 'onboarding', a description, an empty schema object, and the async handler function.
    server.tool(
      "onboarding",
      "Welcome guide for new users. Returns setup instructions, available tools, pricing, and links. No auth required. Call this when someone is new to CI-1T or asks how to get started.",
      {},
      async () => {
        const hasApiKey = !!API_KEY;
        const ready = hasApiKey;
    
        return {
          content: [
            {
              type: "text" as const,
              text: JSON.stringify(
                {
                  welcome: "Welcome to CI-1T — the prediction stability engine.",
                  status: {
                    api_key: hasApiKey ? "✓ configured" : "✗ not set",
                    ready: hasApiKey,
                  },
                  what_is_ci1t:
                    "CI-1T measures whether ML model predictions are stable or collapsing. It computes the Collapse Index (CI) — a Q0.16 fixed-point metric where 0 = perfectly stable and 65535 = total collapse. Useful for detecting silent failures, ghost predictions (confident but wrong), and model drift.",
                  getting_started: [
                    `1. Create a free account at ${SIGNUP_URL} — you get 1,000 free credits on signup`,
                    "2. Go to Dashboard → API Keys → Create a new key (starts with ci_...)",
                    "3. Set CI1T_API_KEY in your MCP client config — that's the only credential you need",
                    "4. Start evaluating — your free credits cover 1,000 episodes",
                  ],
                  environment_variables: {
                    CI1T_API_KEY: {
                      required_for: "All authenticated tools",
                      format: "ci_... (32 chars)",
                      description: "Your API key — single credential for evaluations, fleet sessions, API key management, and billing",
                    },
                    CI1T_BASE_URL: {
                      required_for: "optional",
                      format: "URL",
                      description: `API base URL (default: ${BASE_URL})`,
                    },
                  },
                  config_example: {
                    claude_desktop: {
                      mcpServers: {
                        ci1t: {
                          command: "npx",
                          args: ["-y", "@collapseindex/ci1t-mcp"],
                          env: {
                            CI1T_API_KEY: "ci_your_key_here",
                          },
                        },
                      },
                    },
                  },
                  tools_available_now: [
                    "onboarding — this guide (no auth)",
                    "interpret_scores — statistical breakdown of scores (no auth)",
                    "convert_scores — float ↔ Q0.16 conversion (no auth)",
                    "generate_config — integration boilerplate (no auth)",
                    ...(hasApiKey
                      ? [
                          "evaluate — run stability analysis (API key ✓)",
                          "fleet_evaluate — multi-node fleet analysis (API key ✓)",
                          "probe — test LLM stability (API key ✓)",
                          "health — engine status (API key ✓)",
                          "fleet_session_* — persistent sessions (API key ✓)",
                          "API key management (API key ✓)",
                          "get_invoices — billing (API key ✓)",
                        ]
                      : []),
                  ],
                  pricing: {
                    free_credits: "1,000 credits on signup — no credit card required",
                    evaluate: "1 credit per episode",
                    fleet_evaluate: "1 credit per episode per node",
                    probe: "1 credit per probe",
                    utility_tools: "Free — no credits, no auth",
                  },
                  links: {
                    dashboard: SIGNUP_URL,
                    docs: DOCS_URL,
                    github: "https://github.com/collapseindex",
                  },
                },
                null,
                2
              ),
            },
          ],
        };
      }
    );
  • The requireApiKey() helper function references 'onboarding' as one of the free tools that work without an API key. This reinforces that onboarding does not require authentication.
    function requireApiKey(): McpToolResult | null {
      if (API_KEY) return null;
      return {
        content: [
          {
            type: "text" as const,
            text: JSON.stringify(
              {
                error: "missing_api_key",
                message:
                  "No API key configured. You need a CI-1T API key to use this tool.",
                setup_steps: [
                  `1. Create a free account at ${SIGNUP_URL} — you get 1,000 free credits on signup`,
                  "2. Go to Dashboard → API Keys and create a new key (starts with ci_...)",
                  "3. Set the CI1T_API_KEY environment variable in your MCP client config",
                  "4. Restart the MCP server — all tools will be unlocked",
                ],
                config_example: {
                  claude_desktop: {
                    mcpServers: {
                      ci1t: {
                        env: {
                          CI1T_API_KEY: "ci_your_key_here",
                        },
                      },
                    },
                  },
                },
                free_tools:
                  "You can use interpret_scores, convert_scores, generate_config, and onboarding without an API key or credits.",
              },
              null,
              2
            ),
          },
        ],
      };
    }
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided; description compensates by stating 'No auth required' and describing return content. Adds value beyond schema.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two efficient sentences with no wasted words, front-loaded with purpose and content.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Fully describes purpose, usage, and output for a simple, parameterless tool. No gaps.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Zero parameters, so baseline 4. Description mentions output contents, which is sufficient.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Clearly states it's a welcome guide for new users, listing specific contents (setup instructions, tools, pricing, links). Distinct from sibling tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly says to call when someone is new or asks how to get started, providing clear usage context.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/collapseindex/ci-1t-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server