The Agent Skills Directory

[PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill identifies and executes test commands from project-specific configuration files found in the target repository.
Ingestion points: Reads CLAUDE.md and .ai-cortex/config.yaml from the repository being processed.
Boundary markers: The instructions lack explicit delimiters or warnings to ignore malicious instructions embedded within these configuration files.
Capability inventory: The agent can execute arbitrary shell commands and write to the file system to apply code fixes.
Sanitization: The skill does not describe a process for validating or sanitizing commands parsed from external configuration files before they are executed in the shell.
[COMMAND_EXECUTION]: Automated Command Execution. The skill's primary function involves running shell commands for testing and applying code changes. It explicitly implements safeguards by requiring user confirmation for high-risk operations like network access, Docker usage, or package installation.

automate-repair