skills/aibtcdev/skills/dca/Gen Agent Trust Hub

dca

Pass

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: SAFECREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill accesses sensitive local wallet and keystore files to perform transaction signing. It also supports the use of environment variables for sensitive data.
  • Files accessed: ~/.aibtc/wallets.json, ~/.aibtc/wallets/*/keystore.json, and ~/.aibtc/wallet.json.
  • Environment variables: STACKS_PRIVATE_KEY (direct key access) and AIBTC_WALLET_PASSWORD (keystore decryption).
  • The skill implements best practices by advising against passing passwords via CLI flags and uses AES-256-GCM for keystore decryption.
  • [COMMAND_EXECUTION]: The install-packs command programmatically executes a shell command to manage its environment.
  • Implementation: Uses spawnSync to run bun add for installing its own dependencies.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the download of official blockchain and DeFi SDKs during its setup phase.
  • Source: Fetches standard packages (e.g., @bitflowlabs/core-sdk, @stacks/transactions) from the npm registry.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied token symbols and plan data, creating a surface for indirect instructions.
  • Ingestion points: CLI arguments for token selection and locally stored plan files in ~/.aibtc/dca/.
  • Boundary markers: The agent instructions mandate an explicit user confirmation gate (--confirm) and a visual quote review before any funds are spent.
  • Capability inventory: Ability to sign/broadcast mainnet transactions and execute package installation commands.
  • Sanitization: Validates token symbols against the Bitflow API, enforces a 10% maximum slippage cap, and performs pre-transaction STX balance checks.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 1, 2026, 05:03 AM