The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and transform untrusted data from package changelogs into blog post content.
Ingestion points: The agent is instructed to read NEWS.md, CHANGELOG, or NEWS files as defined in Step 1 of the workflow in SKILL.md.
Boundary markers: There are no explicit instructions to use delimiters or ignore potential commands embedded within the processed NEWS content.
Capability inventory: The skill has the capability to read local files, execute a local R script (scripts/get_contributors.R), and write new markdown files to the repository.
Sanitization: The instructions do not specify any validation or filtering of the ingested content before it is processed by the agent.
[COMMAND_EXECUTION]: The skill utilizes a local R script to automate the retrieval of contributor information.
Evidence: SKILL.md contains instructions to run Rscript scripts/get_contributors.R "owner/repo".
Functionality: The script scripts/get_contributors.R acts as a wrapper for the well-known usethis::use_tidy_thanks() function to fetch metadata from the GitHub API.
[EXTERNAL_DOWNLOADS]: The skill provides instructions for installing dependencies from official R and Python package registries.
Evidence: SKILL.md and references/content-guidelines.md describe using install.packages(), pak::pkg_install(), and pip install to set up the necessary environment.
Sources: The mentioned sources include well-known services and organizations, such as r-lib on GitHub and the standard CRAN/PyPI registries.

release-post