The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill contains instructions for the agent to install the ContHunt CLI by downloading and executing remote scripts from the vendor's domain (conthunt.app). The instructions specify that the agent should install the tool immediately without seeking additional user confirmation.
Evidence:
macOS/Linux: curl -fsSL https://conthunt.app/install.sh | sh in SKILL.md and README.md.
Windows PowerShell: irm https://conthunt.app/install.ps1 | iex in SKILL.md.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from social media platforms (TikTok, Instagram, YouTube) and uses that data to answer user questions, which could lead to instructions in the content influencing the agent's behavior.
Ingestion points: Content is retrieved via conthunt research get, conthunt research evidence, and conthunt search get (referenced in SKILL.md and references/commands.md).
Boundary markers: None detected; instructions do not specify delimiters or warnings for the agent to ignore instructions embedded in the external content.
Capability inventory: The skill can execute various CLI commands and perform workspace file operations such as conthunt files read and conthunt files download (referenced in references/commands.md).
Sanitization: No sanitization or filtering of external content is described.
[COMMAND_EXECUTION]: The skill relies on the execution of the conthunt CLI and npx commands to manage social research, authentication, and updates.
Evidence: npx skills add Synthenova/conthunt-cli, conthunt login, conthunt search, and conthunt research commands throughout the skill files.

conthunt