The Agent Skills Directory

[COMMAND_EXECUTION]: The skill frequently executes shell commands to interact with the local environment and Drupal.org services. This includes using the git CLI for repository management and the drupalorg CLI for issue and merge request orchestration. It also attempts to run vendor/bin/phpcs if available to lint code changes.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) because it ingests untrusted external data that the agent then processes to make decisions.
Ingestion points: The agent reads data from drupalorg issue:show (issue descriptions and comments), drupalorg mr:logs (automated test results/logs), and the local CLAUDE.md configuration file.
Boundary markers: The instructions do not define clear delimiters or include warnings for the agent to ignore instructions embedded within the retrieved issue data or logs.
Capability inventory: The agent has the capability to execute shell commands, modify git remote configurations, write to the local filesystem, and push code to remote repositories.
Sanitization: There is no evidence of sanitization or filtering of the content retrieved from external sources before it is analyzed by the agent. The risk is partially mitigated by the inclusion of [PAUSE] checkpoints requiring user confirmation before significant actions.

drupalorg-work-on-issue