The Agent Skills Directory

[NO_CODE] (SAFE): The skill consists entirely of markdown documentation, rules, and templates. There are no executable scripts, binaries, or active code components provided within the skill files.
[COMMAND_EXECUTION] (SAFE): The skill mentions and suggests development commands (e.g., pnpm dev, wrangler d1) in its templates and maintenance reports. These are intended for the user to execute and are not triggered automatically by the skill.
[EXTERNAL_DOWNLOADS] (SAFE): The /docs-update workflow references npm view to check for the latest package versions. This is a metadata-only query to a trusted public registry and does not involve downloading or executing remote code.
[DATA_EXFILTRATION] (SAFE): The skill reads local files such as .env and package.json to identify variable names and dependencies for documentation purposes. It does not perform any network operations to send this data to external servers.
[PROMPT_INJECTION] (SAFE): Although the skill ingests data from local project files (Ingestion points: package.json, .env, wrangler.jsonc; Boundary markers: Template delimiters used; Capability inventory: Reading files and proposing markdown edits; Sanitization: None explicitly defined), this behavior is limited to the local environment and the intended documentation scope, presenting no significant risk of injection or behavioral override.

docs-workflow