The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill functions by ingesting and processing data from external sources and local files, which is a standard pattern for market analysis tools but creates a surface for indirect prompt injection.
Ingestion points: The skill reads from a local file named app-marketing-context.md and processes output from several data-gathering tools including get_trending_keywords, get_market_movers, and get_featured_apps.
Boundary markers: There are no explicit instructions or delimiters used to separate the ingested data from the skill's operational instructions.
Capability inventory: The skill is limited to data retrieval tool calls and generating text-based reports; it lacks high-risk capabilities such as arbitrary shell execution, file system writes, or direct network requests.
Sanitization: The instructions do not specify any validation or sanitization steps for the data retrieved from external tools or local files.
[SAFE]: The skill's behavior is consistent with its stated purpose. It does not contain any obfuscated code, attempts to access sensitive system files, persistence mechanisms, or unauthorized network operations.

market-pulse