The Agent Skills Directory

[COMMAND_EXECUTION]: The skill verifies the environment by executing shell commands like node -v, npm ci, and supabase status. These actions are necessary to confirm that the required runtimes and services are available.
[EXTERNAL_DOWNLOADS]: The skill confirms dependencies are installed using commands like npm install or pip install. This involves downloading and installing external packages from official registries as part of the setup process.
[DATA_EXFILTRATION]: The skill analyzes .env and configuration files to ensure required variables are present. It contains explicit guards: 'Report only presence/absence — never print, log, or echo secret values' and 'never transmitted', which effectively mitigates the risk of credential exposure.
[PROMPT_INJECTION]: The skill reads external project files such as package.json and Makefile to detect requirements, creating a surface for Indirect Prompt Injection.
Ingestion points: SKILL.md instructs reading of package.json, .env.example, Makefile, README, and other project files.
Boundary markers: Absent. The skill does not explicitly use delimiters or ignore instructions within these files.
Capability inventory: The skill can execute arbitrary shell commands and perform network pings based on discovered config.
Sanitization: None. The skill assumes the project files are legitimate and does not filter or sanitize their content.

workflow-environment-ready