{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"npm_npm-mcpusage","slug":"npm-mcpusage","name":"mcpusage","type":"mcp","url":"https://www.npmjs.com/package/mcpusage","page_url":"https://unfragile.ai/npm-mcpusage","categories":["mcp-servers"],"tags":["mcp","tokenizer","cli"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"npm_npm-mcpusage__cap_0","uri":"capability://data.processing.analysis.mcp.server.tool.advertisement.token.usage.measurement","name":"mcp server tool advertisement token usage measurement","description":"Analyzes MCP (Model Context Protocol) server tool definitions and calculates token consumption for tool advertisement payloads using a built-in or configurable tokenizer. The tool parses tool schemas (name, description, input_schema) and computes tokens consumed when these tools are advertised to LLM clients, enabling developers to understand the cost of exposing tool catalogs in MCP servers.","intents":["Measure how many tokens are consumed when my MCP server advertises its available tools to clients","Understand the token cost impact of adding new tools to my MCP server's tool catalog","Compare token efficiency between different tool schema designs before deployment","Identify which tools in my MCP server consume the most tokens in advertisements"],"best_for":["MCP server developers optimizing token budgets for LLM interactions","Teams building cost-aware AI agent infrastructure with MCP","Developers integrating MCP servers with token-limited LLM APIs (OpenAI, Anthropic, etc.)"],"limitations":["Measures only tool advertisement tokens, not execution or response tokens from actual tool calls","Token counting accuracy depends on the underlying tokenizer implementation — may not perfectly match production LLM tokenization","No built-in support for dynamic tool registration patterns — requires static tool definitions","CLI-only interface; no programmatic API for integration into build pipelines or monitoring systems"],"requires":["Node.js 14+ (typical for npm packages)","Valid MCP server configuration or tool schema definitions","Access to tokenizer (built-in or external dependency)"],"input_types":["MCP server configuration (JSON/YAML)","Tool schema definitions (JSON)","Tool metadata (name, description, input_schema)"],"output_types":["Token count (numeric)","Per-tool token breakdown (structured data)","CLI formatted report (text)"],"categories":["data-processing-analysis","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm_npm-mcpusage__cap_1","uri":"capability://automation.workflow.cli.based.batch.tool.schema.analysis","name":"cli-based batch tool schema analysis","description":"Provides a command-line interface for processing multiple tool schemas or MCP server configurations in batch, computing aggregate and per-tool token metrics. The CLI accepts file paths or stdin input, parses tool definitions, and outputs results in configurable formats (JSON, table, summary), enabling integration into shell scripts and CI/CD pipelines for automated token budget validation.","intents":["Run token analysis on my MCP server configuration as part of CI/CD to catch token budget regressions","Generate a report of token consumption across all tools in my MCP server","Automate token measurement in a shell script or build process","Export tool token metrics in JSON format for downstream analysis or dashboards"],"best_for":["DevOps engineers integrating token cost checks into deployment pipelines","MCP server maintainers monitoring tool catalog growth and token impact","Teams using infrastructure-as-code patterns for MCP server definitions"],"limitations":["CLI output formats are fixed — no custom templating for report generation","No streaming support for very large tool catalogs (memory-bound processing)","Requires manual invocation; no built-in scheduling or continuous monitoring","Error handling and validation feedback may be minimal in early versions"],"requires":["Node.js 14+ with npm","CLI access (terminal/shell environment)","Tool schema files in supported format (JSON/YAML)"],"input_types":["File paths to tool schema definitions","stdin piped tool schema data","MCP server configuration files"],"output_types":["JSON formatted token metrics","Human-readable table/summary output","Exit codes for CI/CD integration"],"categories":["automation-workflow","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm_npm-mcpusage__cap_2","uri":"capability://tool.use.integration.tool.schema.tokenization.with.configurable.tokenizer.backend","name":"tool schema tokenization with configurable tokenizer backend","description":"Abstracts tokenizer implementation to support multiple backend tokenizers (e.g., tiktoken for OpenAI, custom tokenizers for other LLM providers), allowing users to measure token consumption using the same tokenizer their target LLM uses. The tool accepts a tokenizer configuration parameter and applies it consistently across all tool schema analysis, ensuring token counts match production LLM behavior.","intents":["Measure tool tokens using OpenAI's tiktoken to match my production LLM tokenization","Switch tokenizers to test token consumption against different LLM providers (Anthropic, Cohere, etc.)","Ensure my token measurements align with the specific LLM I'm integrating with","Validate that tool schemas fit within my LLM's context window constraints"],"best_for":["Teams using multiple LLM providers and needing provider-specific token counts","Developers building cost-aware systems that need accurate per-provider token budgeting","MCP server operators optimizing for specific LLM backends"],"limitations":["Tokenizer accuracy depends on external library implementations — may lag behind LLM provider updates","No built-in tokenizer for all providers; requires explicit configuration or installation of provider-specific packages","Switching tokenizers requires CLI flag or config change; no automatic detection of target LLM","Custom tokenizer integration requires developer effort if provider not pre-supported"],"requires":["Node.js 14+","Tokenizer library for target provider (e.g., tiktoken for OpenAI)","Tokenizer configuration (name, model identifier, or custom implementation path)"],"input_types":["Tokenizer backend identifier (string)","Tool schema definitions (JSON)","Optional tokenizer configuration parameters"],"output_types":["Token counts (numeric, provider-specific)","Token breakdown by tool component (name, description, schema)"],"categories":["tool-use-integration","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"npm_npm-mcpusage__cap_3","uri":"capability://data.processing.analysis.tool.schema.component.level.token.breakdown","name":"tool schema component-level token breakdown","description":"Decomposes token consumption across individual tool schema components (tool name, description, input_schema, required fields, type definitions) and reports token counts per component. This granular analysis helps developers identify which parts of tool definitions consume the most tokens and where optimization opportunities exist, using a component-aware parsing strategy.","intents":["Identify which tool descriptions are consuming the most tokens and optimize them","Understand the token cost of complex input schemas vs simple ones","Find the most expensive tools in my catalog and prioritize optimization","Compare token efficiency of different schema design patterns (nested objects vs flat structures)"],"best_for":["MCP server developers optimizing tool schemas for token efficiency","Teams with large tool catalogs looking to reduce token overhead","Cost-conscious AI infrastructure teams tracking token consumption per tool"],"limitations":["Component breakdown may not align perfectly with how LLMs tokenize nested JSON structures","No optimization suggestions — only measurement; developers must manually refactor schemas","Complex schema types (recursive definitions, polymorphic types) may not decompose cleanly","Breakdown granularity depends on tokenizer backend capabilities"],"requires":["Node.js 14+","Tool schema definitions with standard components (name, description, input_schema)","Tokenizer backend that supports component-level analysis"],"input_types":["Tool schema JSON with name, description, input_schema fields","MCP tool definition format"],"output_types":["Per-component token counts (structured data)","Ranked list of most expensive components","Comparative metrics (tokens per tool, average description length)"],"categories":["data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":28,"verified":false,"data_access_risk":"high","permissions":["Node.js 14+ (typical for npm packages)","Valid MCP server configuration or tool schema definitions","Access to tokenizer (built-in or external dependency)","Node.js 14+ with npm","CLI access (terminal/shell environment)","Tool schema files in supported format (JSON/YAML)","Node.js 14+","Tokenizer library for target provider (e.g., tiktoken for OpenAI)","Tokenizer configuration (name, model identifier, or custom implementation path)","Tool schema definitions with standard components (name, description, input_schema)"],"failure_modes":["Measures only tool advertisement tokens, not execution or response tokens from actual tool calls","Token counting accuracy depends on the underlying tokenizer implementation — may not perfectly match production LLM tokenization","No built-in support for dynamic tool registration patterns — requires static tool definitions","CLI-only interface; no programmatic API for integration into build pipelines or monitoring systems","CLI output formats are fixed — no custom templating for report generation","No streaming support for very large tool catalogs (memory-bound processing)","Requires manual invocation; no built-in scheduling or continuous monitoring","Error handling and validation feedback may be minimal in early versions","Tokenizer accuracy depends on external library implementations — may lag behind LLM provider updates","No built-in tokenizer for all providers; requires explicit configuration or installation of provider-specific packages","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.05,"quality":0.18,"ecosystem":0.38999999999999996,"match_graph":0.25,"freshness":0.9,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.15,"match_graph":0.23,"freshness":0.12}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:23.904Z","last_scraped_at":"2026-04-22T08:10:07.730Z","last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=npm-mcpusage","compare_url":"https://unfragile.ai/compare?artifact=npm-mcpusage"}},"signature":"1Z95XPgDZkiAPdPCvEDmWCHgztP1UwIe0pcwervZEQOR9VV9CivftgOjaA4xFPsJOVH9v44SPHQIMAQMdHy1Bg==","signedAt":"2026-06-16T05:25:26.365Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/npm-mcpusage","artifact":"https://unfragile.ai/npm-mcpusage","verify":"https://unfragile.ai/api/v1/verify?slug=npm-mcpusage","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}