The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's procedure involves extracting content from project.md and work/*/README.md and interpolating it into shell arguments for the asta documents add command. Specifically, strings for --name and --summary are taken directly from these files. If the processed files contain shell-sensitive characters like backticks, semicolons, or dollar signs, the agent may inadvertently execute arbitrary commands.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection vulnerability surface as it processes external/untrusted data from files without robust sanitization.
Ingestion points: The skill reads project.md and work/*/README.md (Procedure steps 1 and 2).
Boundary markers: No explicit delimiters or boundary markers are used to separate the extracted file content from the shell command template.
Capability inventory: The skill utilizes Bash capabilities for git operations and document indexing.
Sanitization: There are no instructions to sanitize or escape the content extracted from the files before it is passed to the shell environment.

save-work