The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a Python script (scripts/monero_wallet_rpc.py) to manage Monero transactions. This script interacts with a local Monero gateway using the requests library to perform financial operations.\n- [PROMPT_INJECTION]: The skill is vulnerable to Indirect Prompt Injection (Category 8) due to its handling of the XMR402 protocol.\n
Ingestion points: The agent is instructed in SKILL.md to parse WWW-Authenticate headers from external, potentially untrusted web servers.\n
Boundary markers: Absent. The instructions require the agent to strictly follow the challenge parameters provided by the external server.\n
Capability inventory: The skill possesses the capability to perform financial transfers (transfer) and pay protocol challenges (pay-402) via the monero_wallet_rpc.py script.\n
Sanitization: The Python script validates that the amount is a float, but it does not implement destination whitelisting or user-confirmation gates for the address and amount provided by the external source.\n- [DATA_EXFILTRATION]: The skill handles a sensitive AGENT_API_KEY. It employs security best practices by hardcoding the destination URL to 127.0.0.1 and disabling environment-based proxy settings (trust_env=False), which prevents the accidental transmission of credentials to external or attacker-controlled servers.

monero-wallet