The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/vixen_config.py uses subprocess.run to execute the mcporter command-line tool. This is the standard platform interface for Senpi agents to interact with environment services and exchange data.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from market sources.
- Ingestion points: External market data is fetched via the leaderboard_get_markets and market_get_asset_data tools in scripts/vixen-scanner.py.
- Boundary markers: The data is processed and output as JSON without explicit delimiters or instructions to the agent to ignore potentially malicious content within the data fields.
- Capability inventory: The skill has the ability to execute shell commands via the mcporter utility in scripts/vixen_config.py.
- Sanitization: The scripts perform type conversion for numerical values but do not sanitize string identifiers (such as token names or trader descriptions) which could theoretically contain instructions targeting the LLM.

vixen-strategy