The Agent Skills Directory

[SAFE]: The skill implements a restricted workflow that focuses exclusively on documentation. It explicitly forbids modifying source code, running build/test commands, or initiating implementation, effectively limiting its capabilities to file creation and updates within specific documentation directories.
[SAFE]: File system access is scoped to searching and writing Markdown files in pre-defined paths (e.g., docs/ideas/). The use of rg (ripgrep) for indexing existing ideas is a standard, low-risk development operation.
[SAFE]: The skill includes explicit authorization logic, ensuring the agent does not transition a project from the 'idea' phase to 'implementation' without unambiguous user consent.
[SAFE]: No obfuscated content, hardcoded credentials, remote code execution patterns, or suspicious network operations were detected in the instructions or templates.

i-have-an-idea