primitive-chat

Pass

Audited by Gen Agent Trust Hub on Jun 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use the primitive CLI to manage accounts (primitive account show), sign up (primitive agent start-agent-signup), and perform email operations (primitive chat, primitive emails). These commands are part of the intended functionality of the vendor's tool.
  • [EXTERNAL_DOWNLOADS]: The skill requires installing the @primitivedotdev/cli package via npm or a Homebrew tap. These resources are owned and maintained by the skill's author ('primitivedotdev').
  • [PROMPT_INJECTION]: The skill processes inbound emails, which is a known surface for indirect prompt injection (Category 8). Malicious content in a received email could attempt to influence the agent's behavior. The skill documentation mitigates this by instructing the agent to use structured --json output for parsing and providing explicit guidelines on when to seek user confirmation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 11, 2026, 06:22 PM
Security Audit — agent-trust-hub — primitive-chat