The Agent Skills Directory

[NO_CODE]: The skill consists entirely of natural language instructions and markdown formatting. There are no scripts, executables, or code blocks that perform actions on the system.
[SAFE]: The skill does not request or utilize any sensitive permissions, network access, or credentials. All operations described are confined to text analysis and content generation based on user-provided profile information.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data, such as Threads handles and bio text. While this is an ingestion surface for potential indirect injections, the skill lacks any capabilities (such as network calls or shell execution) to manifest a security threat from such input. The risk is limited to the agent potentially being influenced by text contained within a profile bio during the audit process.

threads-profile-optimizer