byreal-cli

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill directs the installation of the '@byreal-io/byreal-cli' package globally from the npm registry. This is the primary functional component of the skill.
  • [DATA_EXFILTRATION]: The skill requires access to the '~/.config/byreal/keys/' directory, which is intended to store sensitive Solana private keys. This access is declared in the metadata to enable local transaction signing by the CLI tool.
  • [COMMAND_EXECUTION]: The skill instructs the agent to run 'byreal-cli' subcommands, including 'update install', which facilitates the retrieval and execution of tool updates.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by instructing the agent to fetch documentation and capability schemas from the output of 'byreal-cli skill' and 'byreal-cli catalog list'. * Ingestion points: Output of 'byreal-cli skill' and 'byreal-cli catalog' commands. * Boundary markers: Absent; there are no instructions or delimiters to help the agent distinguish tool output from trusted system instructions. * Capability inventory: The skill possesses capabilities for on-chain write operations, such as token swaps and liquidity management, via the 'byreal-cli'. * Sanitization: Absent; the agent is instructed to process the raw output of the external tool as its primary documentation source.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 02:07 PM
Security Audit — agent-trust-hub — byreal-cli