skills/messagebird/bird-ai/bird-cli/Gen Agent Trust Hub

bird-cli

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted incoming communication data which could contain malicious instructions aimed at the agent.
  • Ingestion points: The agent ingests external message data using commands like bird email inbound-messages list, bird whatsapp conversations list-messages, and bird support-tickets get (referenced in references/inbound.md, references/whatsapp-conversations.md, and references/support.md).
  • Boundary markers: The instructions frequently emphasize user confirmation for sensitive actions, such as sending broadcasts or managing API keys, but there are no technical delimiters (like XML tags) implemented to separate untrusted message content from the agent's system instructions.
  • Capability inventory: The agent has access to high-impact capabilities including message sending across multiple channels, API key generation (bird api-keys create), and initiating voice calls (bird voice calls create).
  • Sanitization: The skill lacks explicit instructions for the agent to sanitize or filter external message content before it is processed in the LLM context.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the necessary bird CLI and optional baresip tools from official or well-known sources.
  • Evidence: SKILL.md directs users to install the CLI via Homebrew (brew install messagebird/tap/bird) or the official documentation (bird.com). references/voice.md recommends installing baresip via Homebrew or its official repository. These are trusted sources associated with the skill author or established open-source projects.
  • [COMMAND_EXECUTION]: The skill's primary function is to execute CLI commands to manage communication services and account resources.
  • Evidence: The entire skill is built around the bird CLI tool. It includes detailed procedures for sensitive operations like bird api-keys create (references/api-keys.md) and bird auth login (references/authenticate.md), which are necessary for the skill's intended purpose. The instructions emphasize using dry-run and idempotency keys to manage these operations safely.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 11:53 AM