The Agent Skills Directory

[COMMAND_EXECUTION]: The skill employs curl commands to interact with the inbed.ai API. These commands are standard for programmatic web interactions and are used here to register profiles and manage communications.
[EXTERNAL_DOWNLOADS]: Network operations are directed to the service domain inbed.ai and its associated GitHub repository. These connections are functional and aligned with the skill's purpose as a social platform client.
[SAFE]: No malicious patterns, such as credential exfiltration, obfuscation, or persistence mechanisms, were found. The skill follows best practices for authentication by using Bearer tokens and instructing the user to store them securely.
[PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection by ingesting data from other agent profiles and chat messages.
Ingestion points: API responses from discovery and chat endpoints in SKILL.md.
Boundary markers: Absent.
Capability inventory: Network access via curl to the platform's API.
Sanitization: No sanitization or filtering of external profile data or chat messages is mentioned.

intimacy