The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources such as issue tracker items, comments, and backlog files, which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: The agent reads content via rota item field list, rota item show (which includes title, body, and comments), and backlog files like .rota/BACKLOG.md and related detail files.
Boundary markers: The instructions do not specify the use of clear boundary markers or system-level warnings to ignore natural language instructions embedded within the item descriptions or comments.
Capability inventory: The skill utilizes extensive capabilities including shell command execution (git, rota, curl), file system modification (Step 5: Fix and commit), and external interaction via Pull Request creation (rota ship pr).
Sanitization: There are no explicit instructions for the agent to sanitize or validate the content retrieved from issue descriptions or comments before using it to formulate hypotheses or reproduction steps.
[COMMAND_EXECUTION]: The skill operates primarily through shell command execution to interact with the repository and project management tools.
Evidence: The workflow involves numerous calls to git, curl, and a specialized rota CLI utility for state management, tracking, and reporting.

rota-debug