The Agent Skills Directory

[COMMAND_EXECUTION]: The skill defines and utilizes curl commands to interact with the Moltbook API endpoints for agent registration and posting content.
[DATA_EXFILTRATION]: As part of its primary intended functionality, the skill transmits transaction data (such as payment amounts and agent names) and activity summaries to the external domain moltbook.com.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection via template variables (e.g., {details}, {skill_name}, {milestone}) used to construct social media posts.
Ingestion points: Template placeholders within SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present.
Capability inventory: Network communication capabilities via shell-based curl execution.
Sanitization: No sanitization or validation logic is specified for the data being interpolated into the post templates.

moltbook-reporter