The Agent Skills Directory

[SAFE]: The skill establishes a logical framework for managing startup projects, focusing on brainstorm, validation, and research phases. No malicious patterns, obfuscation, or unauthorized data access were identified.
[DATA_EXPOSURE]: The skill manages project-specific metadata in a config.json file and organizes project documentation in local directories. There are no instructions to access sensitive system files, environment variables, or credentials.
[PROMPT_INJECTION]: The instructions emphasize an 'Honest feedback' interaction style and strict adherence to 'Stage Gates.' There are no attempts to bypass safety filters or override the agent's core behavioral guidelines.
[COMMAND_EXECUTION]: The skill uses allowed tools for web searching and reading files. It does not contain any shell commands, subprocess executions, or privilege escalation attempts.

startup-pipeline