The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's publishing logic, primarily contained in scripts/publish.sh, utilizes standard command-line utilities such as xurl, ffmpeg, magick (ImageMagick), git, gh, and curl. these tools are employed to process images, render video, and interact with platform APIs, which is consistent with the skill's stated purpose of content distribution.
[EXTERNAL_DOWNLOADS]: The skill makes network requests to the official REST APIs of well-known services, including LinkedIn (api.linkedin.com), Instagram (graph.instagram.com), Dev.to, and Hashnode. These operations are used for legitimate content publishing and target established service providers.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data from the web through compounding skills like /agent-browser and /deep-research to inform its content generation. Ingestion points: External URLs and research data extracted during Phase 1. Boundary markers: Absent, as there are no specific delimiters used to isolate processed research data from the agent's internal instructions. Capability inventory: Includes file system writes, execution of various CLI tools, and network operations to social media APIs. Sanitization: No explicit sanitization or validation of the ingested data is described before it is used to generate published content.

blog-post