The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes and acts upon data from an external CI service.\n
Ingestion points: Data enters the agent's context through the ci-monitor-subagent and ci_information tool, which provide suggestedFix and taskOutputSummary from Nx Cloud.\n
Boundary markers: The skill does not use specific delimiters to isolate external fix suggestions from the agent's internal logic.\n
Capability inventory: The skill can execute shell commands (Bash), perform git operations (commit/push), and run build tasks (nx run).\n
Sanitization: The skill lacks validation for suggested code patches before they are applied to the local filesystem or committed to the repository.\n- [COMMAND_EXECUTION]: The skill executes standard development commands to monitor and maintain the project state.\n
Evidence: Uses git for branch detection and version control operations. Executes nx run to locally verify the results of CI tasks.\n- [EXTERNAL_DOWNLOADS]: The skill uses established package managers to update project dependencies.\n
Evidence: Runs pnpm install, npm install, or yarn install to update lockfiles as part of the auto-fix workflow.\n- [REMOTE_CODE_EXECUTION]: The skill fetches and applies code modifications from the Nx Cloud service.\n
Evidence: Utilizes nx-cloud apply-locally to download patches from Nx Cloud and apply them to the local workspace for verification.

monitor-ci