The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown instructions and templates for competitive research. No executable scripts, binaries, or configuration files that execute code are included in the skill package.
[SAFE]: The skill instructs the agent to read a local configuration file (.agents/product-marketing-context.md) to obtain business context. This is a standard mechanism for agent initialization and does not involve accessing sensitive system directories.
[EXTERNAL_DOWNLOADS]: The research workflow involves fetching content from external websites and review platforms. 1. Ingestion points: Competitor websites and review pages (G2, Capterra) accessed during the research phase. 2. Boundary markers: Absent. 3. Capability inventory: The skill lacks capabilities for file system modification, command execution, or network exfiltration; its primary output is the synthesis of markdown reports. 4. Sanitization: Absent. The lack of dangerous capabilities effectively mitigates the risks associated with indirect prompt injection from untrusted external content encountered during the research process.

competitor-analysis