The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes data from external, untrusted sources which could contain malicious instructions designed to manipulate the agent's trading logic.
Ingestion points: scripts/news.mjs (news details and lists) and scripts/twitter.mjs (tweet content and search results).
Boundary markers: No specific delimiters or instructions are present to prevent the agent from following instructions embedded within the fetched content.
Capability inventory: The agent has the ability to place orders, set leverage, and modify bot configurations via scripts/exchange.mjs and scripts/auto-trade.mjs.
Sanitization: Content is fetched and displayed without sanitization or escaping of potential injection patterns.
[COMMAND_EXECUTION]: Several scripts utilize execSync to perform infrastructure management and execution tasks.
scripts/ft-deploy.mjs executes shell commands to install Python, clone the Freqtrade repository, run its setup script, and manage background processes.
scripts/auto-trade.mjs uses subprocesses to call the internal exchange.mjs script for executing trading operations.
[EXTERNAL_DOWNLOADS]: scripts/ft-deploy.mjs clones the Freqtrade source code from its official GitHub repository. This is a well-known technology service, and the operation is a standard part of the documented deployment process for the trading bot.
[CREDENTIALS_UNSAFE]: The skill manages authentication for multiple services through local environment files.
It automatically loads .env files from the workspace and home directories to retrieve exchange API keys.
lib/defaults.json includes a hardcoded public free-tier access key and secret for the AiCoin API, which serves as a fallback for users without personal credentials.

aicoin