The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell scripts to perform lifecycle management, including copying internal scripts to a hidden .trogteam/ directory and modifying file permissions using chmod +x before execution.
[COMMAND_EXECUTION]: The script spawn-agents.sh dynamically builds and executes commands in new terminal windows using osascript on macOS and gnome-terminal or xterm on Linux systems.
[REMOTE_CODE_EXECUTION]: The skill utilizes opencode serve and opencode run --attach to create a local execution environment, which acts as a mechanism for dynamic code execution during agent loops.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) through its ingestion of external content.
Ingestion points: The skill reads task metadata, descriptions, and comments from the bd CLI tool and project-specific guidelines from GUARDRAILS.md.
Boundary markers: No delimiters or protective instructions are used to separate ingested data from the agent's core instructions.
Capability inventory: The skill has the capability to perform git operations (fetch, merge, push), modify the file system, and spawn background processes.
Sanitization: There is no evidence of sanitization or validation of the data retrieved from external task lists or local files before it is interpolated into the agent's context.

grug