The Agent Skills Directory

[SAFE]: The skill demonstrates secure credential management by instructing the agent to read the BESTBLOGS_API_KEY from an environment variable rather than requiring it to be hardcoded.
[COMMAND_EXECUTION]: The skill uses curl to interact with the BestBlogs.dev OpenAPI. The commands are structured using defined JSON payloads and environment variables, minimizing the risk of arbitrary command injection through user-controlled parameters.
[PROMPT_INJECTION]: The skill processes external data, which constitutes a surface for indirect prompt injection.
Ingestion points: The skill fetches article titles, summaries, key quotes, and full Markdown content from the api.bestblogs.dev endpoint (documented in SKILL.md and references/api_reference.md).
Boundary markers: There are no explicit delimiters or instructions provided in the output templates to help the model distinguish between fetched content and system instructions.
Capability inventory: The skill's capabilities are limited to network requests via curl and formatting Markdown output; it does not possess capabilities for file-system modification or secondary shell execution.
Sanitization: No sanitization or filtering logic is described for the content retrieved from the external API before it is rendered to the user or the agent's context.

bestblogs-fetcher