The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx agentic-flow@alpha in both pre_execution and post_execution hooks. This fetches and executes the latest alpha version of the agentic-flow package from the NPM registry at runtime.
[COMMAND_EXECUTION]: The skill defines pre_execution and post_execution hooks that execute shell commands, including environment checks (npx --version) and memory pattern storage. This is a standard feature for this agent platform but involves runtime command execution.
[INDIRECT_PROMPT_INJECTION]: The post_execution hook contains a potential command injection surface by interpolating the $TASK variable directly into a shell command.
Ingestion points: The $TASK variable is used within the post_execution hook in SKILL.md.
Boundary markers: The variable is enclosed in double quotes ("Memory Unification: $TASK"), which offers minimal protection against shell metacharacters.
Capability inventory: The hook has the capability to execute shell commands via npx.
Sanitization: There is no evidence of sanitization or escaping applied to the $TASK variable before it is passed to the shell command.

agent-v3-memory-specialist