The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Node.js script located at ~/.codex/supermemory/save-memory.js. This is a vendor-owned resource managed by 'supermemoryai' and is required for the skill's primary function of persisting knowledge.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection that could lead to command injection because it interpolates untrusted conversation content directly into a bash command string.
Ingestion points: The agent gathers project knowledge and user goals from the conversation history as described in SKILL.md.
Boundary markers: The skill uses [SAVE] and [/SAVE] container tags to structure the data, but these do not provide security boundaries against shell breakout.
Capability inventory: The skill is granted access to the Bash tool to execute node commands.
Sanitization: The instructions lack requirements for the agent to escape or sanitize the FORMATTED_CONTENT string, which could allow a malicious user to execute arbitrary commands by including shell metacharacters in their request.

supermemory-save