The Agent Skills Directory

[SAFE]: The skill's primary function is to scaffold and maintain documentation. All operations are confined to the project directory and involve standard markdown file management.
[COMMAND_EXECUTION]: The skill uses commands to detect project structure, create directories (mkdir), and write markdown files. These are benign and necessary for the documentation workflow.
[DATA_EXFILTRATION]: No network activity was detected. The skill does not attempt to contact external servers or exfiltrate any data. It focuses on reading local configuration files like CLAUDE.md or AGENTS.md for orientation purposes.
[PROMPT_INJECTION]: The instructions are designed to provide operational context to the agent. There are no attempts to bypass safety filters, extract system prompts, or override the agent's core safety guidelines.
[INDIRECT_PROMPT_INJECTION]: While the skill reads existing project files (CLAUDE.md, CHANGELOG.md), which is a common surface for indirect injection, it uses these for context rather than execution, and the suggested integration methods include delimiters to help separate documentation from instructions.

project-chronicles