The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the curl utility to perform network requests to the multimodal API. User-provided prompts and media URLs are included in the request body of these commands.
[DATA_EXFILTRATION]: Network activity is directed to https://api.jiekou.ai, which is the official API endpoint for the skill's vendor. User-supplied data is transmitted to this domain as part of the skill's core functionality.
[PROMPT_INJECTION]: The skill processes untrusted user content, such as text prompts and media URLs, representing an indirect prompt injection surface.
Ingestion points: User input enters the skill context through prompts and URLs provided for multimodal tasks described in SKILL.md and references/examples.md.
Boundary markers: Absent. The instructions do not specify the use of delimiters or isolation techniques for user-provided data.
Capability inventory: The skill performs network operations via curl and reads configuration data from the local file system at ~/.jiekou/config.json.
Sanitization: Absent. No input validation or sanitization routines are defined for the data processed by the skill.

jiekou-multimodal