The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill specifies the installation of a third-party command-line utility, summarize, through the steipete/tap/summarize Homebrew repository.
[COMMAND_EXECUTION]: The instructions direct the agent to execute the summarize binary, passing user-provided URLs and local file paths (such as PDFs) as arguments.
[DATA_EXFILTRATION]: The skill handles sensitive AI provider API keys (OpenAI, Anthropic, Google, xAI) via environment variables and interacts with external services such as Firecrawl and Apify for content extraction. This behavior is consistent with the skill's primary purpose of summarizing external content using LLMs.
[PROMPT_INJECTION]: The skill processes untrusted data from external sources (web pages, transcripts, local files) which presents an attack surface for indirect prompt injection.
Ingestion points: External URLs, YouTube video links, and local file paths (e.g., /path/to/file.pdf).
Boundary markers: None are specified in the instructions to separate content from instructions.
Capability inventory: The skill is capable of executing the summarize CLI tool and accessing the file system to read local files.
Sanitization: The instructions do not provide explicit methods for sanitizing the ingested content before it is processed by the agent.

summarize