skills/nylas/skills/nylas-cli/Gen Agent Trust Hub

nylas-cli

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing the Nylas CLI using Homebrew (nylas/nylas-cli/nylas) or Go (github.com/nylas/cli/cmd/nylas@latest). These sources are official vendor-controlled repositories.
  • [COMMAND_EXECUTION]: The skill enables the agent to execute a wide variety of CLI commands, including nylas email send, nylas calendar events create, and nylas auth token, which displays API credentials. It also includes nylas webhook server, which can initiate a local server and a cloudflared tunnel.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides functions to read and analyze untrusted external data from emails and calendar threads (nylas email read, nylas email ai analyze).
  • Ingestion points: Email bodies, thread content, and meeting links processed via CLI commands in rules/email-commands.md, rules/calendar-commands.md, and rules/notetaker-commands.md.
  • Boundary markers: The instructions do not specify the use of delimiters or specific markers to isolate untrusted data when reading emails.
  • Capability inventory: The skill has the ability to send emails, manage agent accounts, and retrieve OTP codes.
  • Sanitization: No specific sanitization or filtering of external content is mentioned before the data is processed by the agent.
  • [DATA_EXFILTRATION]: The nylas otp get command allows the agent to retrieve 2FA codes from the user's email, which are copied to the system clipboard by default. While this is a documented feature of the tool, it presents a vector for accessing highly sensitive authentication data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 07:13 AM
Security Audit — agent-trust-hub — nylas-cli