The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documentation includes regional API endpoints that are currently flagged as malicious or blacklisted by automated security scanners (e.g., https://api.bybit.id and https://api.bybit.kz).- [METADATA_POISONING]: The skill's metadata claims the author is 'Bybit', which conflicts with the provided author context of 'johnnywic'. This impersonation suggests a high risk of deception intended to gain trust for a skill that requests high-privilege financial access.- [DATA_EXFILTRATION]: The skill provides the agent with capabilities for financial 'Withdraw' (/v5/asset/withdraw/create), 'Internal Transfer', and 'Universal Transfer'. Combined with the agent's access to API keys and secrets, this provides a direct mechanism for unauthorized exfiltration of funds from a user's account.- [COMMAND_EXECUTION]: The skill provides bash script examples using 'curl' and 'openssl' to sign and transmit requests. This instructs the agent to perform sensitive financial operations through system-level shell execution, increasing the attack surface.- [CREDENTIALS_UNSAFE]: The skill requires the agent to manage raw API keys and Secrets, suggesting storage in a local 'TOOLS.md' file. While masking is advised for display, the skill's operational logic requires the agent to process these plaintext secrets programmatically to generate HMAC-SHA256 signatures.- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted JSON data from external API responses, which could potentially contain malicious payloads designed to manipulate the agent's subsequent actions.
Ingestion points: External data enters the agent context through API responses for market data, order history, and account information (SKILL.md).
Boundary markers: There are no specified delimiters or instructions to treat data from the Bybit API as untrusted or to ignore embedded instructions (SKILL.md).
Capability inventory: The skill includes high-impact capabilities such as placing trades, modifying account settings, and initiating withdrawals through system commands (SKILL.md).
Sanitization: The documentation lacks instructions for validating or sanitizing API responses before the agent processes the content for decision-making.

bybit-v5