The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses curl commands to fetch markdown content from https://fullstackrecipes.com/api/recipes/. These external resources are used to guide the agent's setup process for environment variables, database connections, logging, and feature flags.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (curl) to retrieve remote content. While these specific commands only fetch text, the pattern of fetching and then following remote instructions is a known risk vector.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from an external API and expects the agent to act on it.
Ingestion points: External markdown files fetched via curl from fullstackrecipes.com in SKILL.md.
Boundary markers: None. There are no delimiters or warnings to the agent to treat the fetched content as data rather than instructions.
Capability inventory: The skill uses curl for network access to pull setup recipes.
Sanitization: None. The fetched content is processed directly by the agent.

stripe-subscriptions