The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill initiates downloads of external R packages from GitHub repositories (wjschne/apaquarto and crsh/papaja) using the remotes package.
[REMOTE_CODE_EXECUTION]: The skill executes code downloaded from the internet. The command remotes::install_github("wjschne/apaquarto") and remotes::install_github("crsh/papaja") downloads and installs arbitrary code from public GitHub repositories into the execution environment, which is subsequently loadable via library().
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, creating an attack surface for indirect prompt injection.
Ingestion points: Reads and processes user-provided .bib bibliography files and analysis data frames (referenced in Step 5 and Inputs).
Boundary markers: There are no markers or instructions to the agent to ignore embedded commands within the .bib or data files.
Capability inventory: Uses Bash via quarto render to compile documents, Write to create .qmd and .bib files, and remotes to install external code.
Sanitization: The skill lacks sanitization, validation, or escaping of content within the bibliography or analysis data before it is rendered into the final manuscript.

format-apa-report