The Agent Skills Directory

[SAFE]: The skill retrieves financial fundamentals and market data for specified companies. This data access is restricted to the methods described in the skill's context and is consistent with the skill's purpose of financial analysis. The ticker input is used in the output filename (reports/{TICKER}_comps.html), which assumes the execution environment handles path sanitization for the ticker argument.
[COMMAND_EXECUTION]: The skill uses established data retrieval tools and does not execute arbitrary shell commands or system-level operations.
[REMOTE_CODE_EXECUTION]: No patterns of downloading or executing remote code were identified. External references are limited to legitimate citations on the vendor's own domain.
[PROMPT_INJECTION]: The instructions follow a structured analytical workflow and do not contain attempts to bypass agent safety guidelines or override system behavior. The skill ingests company data and user-provided tickers to generate an HTML report. While this creates a theoretical indirect prompt injection surface (e.g., via malicious strings in company metadata), the risk is mitigated by the use of managed financial data sources.
Ingestion points: TICKER argument and company data from discovery tools.
Boundary markers: None explicitly defined in the HTML template.
Capability inventory: File write operations in the reports/ directory.
Sanitization: Not explicitly defined in the prompt text, relying on source data integrity.

comps