The Agent Skills Directory

[SAFE]: The skill operates as a high-level router and coordinator ('Supervisor') for organic marketing tasks. It manages logic for delegating work to specialized sub-modules like afa-seo and afa-social based on user intent.
[DATA_EXFILTRATION]: No unauthorized network operations or data exfiltration patterns were found. The skill processes project-specific data (e.g., keyword lists, brand profiles) within a defined local directory structure (./brand-brain/) and uses YAML for internal agent communication.
[COMMAND_EXECUTION]: The file does not contain any shell commands, subprocess calls, or scripts that execute arbitrary code. It utilizes a declarative routing table and structured workflows to manage agent behavior.
[PROMPT_INJECTION]: The instructions reinforce safety and operational integrity by explicitly forbidding the exposure of internal system codes or module IDs to the end user. It relies on a set of global 'Iron Rules' and system protocols (_system/) for safety enforcement.
[INDIRECT_PROMPT_INJECTION]: The skill manages a 'user_request' field which acts as an entry point for untrusted data. However, the risk is mitigated by the modular design and the reliance on a global protocol framework (_system/preamble.md) for input handling. This represents a standard functional surface with no evidence of exploitable capabilities in this module.

afa-organic