The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches external code from a remote repository.\n
Evidence: git clone https://github.com/23blocks-OS/ai-maestro-plugins.git found in SKILL.md.\n- [REMOTE_CODE_EXECUTION]: The skill executes a script downloaded from an external source.\n
Evidence: ./install-agent-cli.sh execution following the git clone command in SKILL.md.\n- [COMMAND_EXECUTION]: The skill makes extensive use of local shell scripts to perform sensitive operations.\n
Evidence: Commands such as aimaestro-agent.sh create, aimaestro-agent.sh delete, and aimaestro-agent.sh plugin install are used for agent management.\n- [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface where untrusted data could influence agent behavior.\n
Ingestion points: Directory paths (--dir), task descriptions (--task), configuration files (import <file>), and plugin marketplace sources (marketplace add <source>) specified in SKILL.md.\n
Boundary markers: Absent. No delimiters or instructions to ignore embedded commands are present.\n
Capability inventory: File system modification, shell script execution, and plugin management via aimaestro-agent.sh across all documented commands.\n
Sanitization: Absent. No evidence of input validation or content filtering for user-provided data.

agent-management