The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses node to execute local scripts for protocol interaction, including viewing accounts and managing stability pool positions.
[PROMPT_INJECTION]: The skill processes external blockchain data, creating an indirect prompt injection surface.
Ingestion points: Data fetched via getStabilityAccounts and getStabilityPools in scripts/my-accounts.js and scripts/pools.js respectively.
Boundary markers: None identified; external data is printed directly to the console.
Capability inventory: Execution of transaction management scripts as documented in SKILL.md.
Sanitization: Data is logged to the console without explicit filtering or escaping.

indigo-stability