The Agent Skills Directory

[SAFE]: The skill operates entirely through instructions and structured research templates without incorporating malicious code, prompt overrides, or safety bypasses.\n- [COMMAND_EXECUTION]: The use of the Agent tool for parallelization and WebSearch for information gathering is consistent with the skill's documented purpose. These platform-provided tools are used for task delegation and data retrieval, not for executing arbitrary shell commands or unauthorized system operations.\n- [DATA_EXPOSURE]: The skill reads local project files (e.g., competitor-landscape.md, brief.md) to establish context for its research waves. This access is limited to the user's project directory and does not involve harvesting sensitive system credentials or private configuration files.\n- [SAFE]: While the skill ingests data from external sources like forums and review sites, the 'Synthesis Protocol' acts as a security boundary by requiring cross-referencing and confidence labeling of all external findings before they are included in final reports.

startup-competitors