The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocesses to automate data extraction tasks.
Step 2A uses a python3 heredoc to parse the local docs/learning-graph/book-metrics.json file.
Step 2B utilizes a shell pipeline (curl | grep | sed) to extract URL locations from an XML sitemap.
[EXTERNAL_DOWNLOADS]: The skill performs network requests to external domains provided by the user (or templates) using curl. While intended to fetch sitemaps for content discovery, this involves connecting to non-whitelisted URLs.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) due to its data ingestion patterns.
Ingestion points: Untrusted data is retrieved from external websites and sitemaps via curl, as well as from local repository files like mkdocs.yml.
Boundary markers: The skill does not employ explicit delimiters or instructions to ignore malicious content within the fetched data.
Capability inventory: The environment allows execution of shell commands (curl, grep, sed) and python3 scripts.
Sanitization: There is no logic present to sanitize or validate the content of the retrieved sitemaps or web pages before they are used to influence the agent's output.

press-release-generator