The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script, scripts/validate_spec.py, to perform structural and consistency checks on the project specifications. A manual review of this script confirms it is restricted to reading the specified file, performing regular expression matching for tag validation, and outputting findings to the terminal. It does not execute arbitrary code or interact with the system in a dangerous manner.
[DATA_EXFILTRATION]: There are no network-related commands (e.g., curl, wget) or data exfiltration patterns. While the skill processes project details, all operations are local, and the generated output is saved to the user's current directory as intended.
[CREDENTIALS_UNSAFE]: The skill demonstrates safe credential management practices. In the provided examples (e.g., references/example-api-spec.md), it explicitly uses placeholders for sensitive values like JWT keys and instructs users to define secrets via environment variables in an .env file rather than hardcoding them.
[SAFE]: The skill provides clear, instructional content focused on its primary purpose of technical documentation. It does not attempt to bypass safety filters or override agent behavior beyond its defined role as a specification writer.

project-spec-writer