Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only CoinMarketCap API reference with disclosed API-key use, but users should handle keys carefully and treat trading/social signals as informational only.
Install only if you intend to use CoinMarketCap API documentation. Provide API keys through secure environment variables or a secrets manager, avoid exposing account usage details unless needed, and do not treat community posts, sentiment, or fear/greed examples as trading advice.
The skill manifest says this skill is for market-wide CoinMarketCap endpoints, but the referenced file documents content/news/community endpoints instead. This mismatch can cause an agent to select and use the wrong skill, leading to incorrect API calls and exposing downstream workflows to untrusted user-generated content when the caller expected market metrics data.
The file documents Community API endpoints even though the skill metadata says the skill is for market-wide endpoints such as global metrics, fear/greed, indices, trending topics, and charts. This scope drift can cause the agent to answer with unsupported or unintended capabilities, increasing the chance of policy bypass, misuse of the wrong API surface, or accidental disclosure/use of adjacent features.
The use cases explicitly frame community sentiment and trending data as inputs for trading signals, momentum detection, and hype/FOMO monitoring without any caution that such signals are noisy, manipulable, and not investment advice. In an agent context, this can encourage risky financial decision support and over-trust in socially driven indicators, which are especially susceptible to coordinated manipulation.
The manifest says this skill is the complete reference for CoinMarketCap market-wide endpoints such as global metrics, fear/greed, indices, trending topics, and charts. This file instead documents utility endpoints for fiat mapping, API key information, and price conversion, which are outside that stated market-wide scope.
The API key info endpoint exposes account usage, plan, and rate-limit details that are outside the stated market/sentiment purpose of the skill. In an agent setting, broadening a market-data skill to include account-monitoring capabilities can increase the chance of unnecessary access to sensitive operational metadata and can facilitate reconnaissance about API quotas and account state.
The manifest describes this skill as an API reference for market-wide endpoints such as global metrics, fear/greed, indices, trending topics, and charts. This document additionally covers latest crypto news, currency conversion, API key usage info, and fiat currency ID lookup, which are separate content, tools, account, and reference capabilities not indicated by that stated scope.
This guide includes the /v1/key/info endpoint, which exposes account-level API usage and quota details unrelated to the stated market-data purpose. In an agent setting, unnecessary documentation of account-inspection capabilities increases the chance the agent will query sensitive operational metadata, revealing subscription details, rate limits, or usage patterns that can aid reconnaissance or abuse planning.
The documentation describes retrieval of comments and user profile fields from community posts without clearly warning that this is user-generated, potentially adversarial content. In agent contexts, omission of that trust-boundary warning can lead developers to treat comments, usernames, or profile metadata as reliable inputs, increasing prompt-injection, misinformation, or privacy-handling risk.
The use-cases encourage using community posts, comments, and engagement spikes as trading or sentiment signals without cautioning about manipulation, brigading, spam, or low-quality signals. In this skill context, that is more dangerous because the content is explicitly community-generated and can be gamed to influence automated decisions or mislead users.
The file provides actionable trading and risk-management suggestions based on the Fear and Greed Index without an explicit disclaimer that the material is informational and not financial advice. In a market-data skill, users may reasonably act on this guidance as investment advice, increasing the risk of harmful financial decisions despite the content being framed as reference material.
This markdown file includes a concrete example that requires supplying X-CMC_PRO_API_KEY, which involves use of a credential. The surrounding documentation does not include any warning or note about protecting the key, avoiding hard-coding it, or keeping it out of shared logs/history.
The latest-metrics example again instructs users to send an X-CMC_PRO_API_KEY header, but the document provides no user-facing warning about sensitive credential handling. For markdown files, omission of warnings around behaviors affecting privacy or sensitive data qualifies as a missing user warning.
The use-case guidance explicitly encourages API usage monitoring within a market-data reference, normalizing access to account-level telemetry that is not needed for core market or sentiment queries. While not directly exposing the API key itself, this can lead agents to request or reveal billing/rate-limit information beyond the user's intended task, expanding sensitive-data handling unnecessarily.
This markdown file includes repeated curl examples that send X-CMC_PRO_API_KEY but provides no user-facing warning that the key is sensitive credential material. For markdown files, missing warnings about behaviors affecting privacy or credential safety are in scope, and users may copy these examples into shared shells, scripts, or logs without realizing the exposure risk.
