The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill connects to a vendor-controlled external API (https://signals.x70.ai) to fetch trading signals and performance statistics. This behavior is the core functionality and is consistent with the skill's documented purpose.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external API responses, including free-text analysis fields such as summary and tags, which represents a surface for indirect prompt injection.
Ingestion points: Data returned from /api/skill/signals and /api/skill/stats as specified in SKILL.md and references/SIGNAL-FORMAT.md.
Boundary markers: Absent; the instructions do not define specific delimiters for separating external signal content from agent instructions.
Capability inventory: Data retrieval via MCP transport; no local file-writing, command execution, or subprocess capabilities are present.
Sanitization: Absent; the skill directs the agent to format raw signal data into markdown tables for presentation to the user.
[SAFE]: Analysis of the instructions and metadata identifies no evidence of obfuscation, hardcoded credentials, or persistence mechanisms. All network activity is directed to legitimate vendor infrastructure, and placeholder values are used for API key documentation.

trading-signals