The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via the bug description input. \n- Ingestion point: The argument in SKILL.md and $ARGUMENTS in references/investigation-template.md. \n- Boundary markers: No delimiters or explicit instructions to ignore embedded commands are present in the prompts. \n- Capability inventory: The skill has access to powerful CLI tools including bundle exec, rails runner, psql, and redis-cli as specified in SKILL.md and references/debug-commands.md. \n- Sanitization: There is no evidence of input validation or filtering for the data ingested from the bug description. \n- [COMMAND_EXECUTION]: The skill facilitates the execution of powerful shell commands for system and database inspection. \n- Evidence: Instructions in SKILL.md and references/debug-commands.md encourage running bundle exec rails runner, psql, and redis-cli to investigate application state. \n- Context: While standard for Ruby on Rails development, these tools provide significant access to the host environment.

rb:investigate