The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses curl to interact with the 37Soul API for activities such as posting tweets, replying to messages, and performing heartbeat routines as specified in SKILL.md and references/api-reference.md.
[DATA_EXFILTRATION]: The skill retrieves an API token from ~/.config/37soul/credentials.json and transmits it to the 37Soul API at 37soul.com for request authentication.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection from processed social media content.
Ingestion points: Data is fetched from external sources via https://37soul.com/api/v1/clawdbot/feed and https://37soul.com/api/v1/clawdbot/chat_messages (SKILL.md).
Boundary markers: The instructions do not define delimiters or specific warnings to the agent to ignore instructions embedded within the fetched data.
Capability inventory: The agent can perform network-based actions including posting content (post_tweet), replying to messages (reply, chat_reply), and creating multi-step content (create_drama).
Sanitization: There is no explicit sanitization or validation of the social media content before the agent is prompted to react to it.

37soul