The Agent Skills Directory

[COMMAND_EXECUTION]: The skill frequently executes the easykol CLI tool to perform creator discovery, audience analysis, and metadata retrieval. It also uses printf to securely pipe API keys into the authentication command.
[EXTERNAL_DOWNLOADS]: The skill is instructed to install and update the @easykol/cli package from the NPM registry at the start of every session. This package and the associated easykol.com domains belong to the skill's author, iftechio.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from social media profiles (YouTube, TikTok, Instagram) and video metadata, which could contain malicious instructions designed to influence the agent's output.
Ingestion points: Social media profile data, nicknames, and video metadata returned by commands such as easykol kol, easykol audience, and easykol video (SKILL.md, references/platform-support.md).
Boundary markers: The instructions recommend reporting results in plain language but do not define specific delimiters or "ignore instructions" wrappers for the ingested content.
Capability inventory: The skill has the ability to execute shell commands, perform network operations via the vendor's CLI, and write to local task journals in ~/.easykol/.
Sanitization: There are no explicit instructions for sanitizing or filtering the content retrieved from external social media platforms before the agent processes it.

easykol