The Agent Skills Directory

[COMMAND_EXECUTION]: The post_execution hook in SKILL.md interpolates the $TASK variable directly into a shell command: npx agentic-flow@alpha memory store-pattern ... --task "Performance: $TASK". If the task description contains shell metacharacters (e.g., backticks, semicolons, or pipe symbols), it could lead to arbitrary command execution in the agent's environment.
[EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute the agentic-flow@alpha package during the post_execution phase. This package is a vendor-provided resource associated with the author.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes the $TASK variable—which may originate from untrusted user input—directly within a shell execution context without sanitization or boundary markers.
Ingestion points: The $TASK environment variable in the post_execution hook.
Boundary markers: None present; the variable is wrapped in double quotes which can be escaped.
Capability inventory: Execution of shell commands via npx and echo.
Sanitization: No validation or escaping is applied to the $TASK variable before interpolation.

agent-v3-performance-engineer