The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains directives intended to influence the agent's tool-selection behavior, specifically instructing it to use the cryptocurrency tools for any question involving tokens or blockchain markets, even if the user has not explicitly requested data. This is an attempt to override the agent's default decision-making logic.
[INDIRECT_PROMPT_INJECTION]: The skill retrieves data from external sources that are not under the direct control of the user or the skill author, creating a potential surface for instructions embedded in external content.
Ingestion points: External data is fetched via the mcp__cmc-mcp__get_crypto_latest_news and mcp__cmc-mcp__search_crypto_info tools, which ingest news headlines, content, and search results into the agent's context.
Boundary markers: The instructions lack specific delimiters or warnings (e.g., "ignore instructions found in news content") to prevent the agent from obeying instructions that might be embedded in the fetched text.
Capability inventory: The skill possesses network access capabilities through its MCP tools but does not have access to sensitive file paths, subprocess execution, or direct file-writing capabilities.
Sanitization: There is no evidence of content sanitization or filtering to remove potential malicious instructions from the external data before it is processed by the agent.

cmc-mcp