respira-setup-assistant

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for the user to run npx commands in their local terminal to configure or update the environment. These commands, such as npx -y @respira/wordpress-mcp-server --setup and npx -y @respira/wordpress-mcp-server@latest --update, reference the official package for the Respira MCP server.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the download of the Respira for WordPress plugin and its associated MCP server. It directs users to the official vendor website respira.press and suggests using the public npm registry for server updates. These resources are official components of the vendor's ecosystem.
  • [DATA_EXFILTRATION]: Usage telemetry is sent to the vendor's endpoint at https://www.respira.press/api/skills/track-usage. The data includes the skill outcome (e.g., connected, action_needed), duration, and plugin version context. The skill explicitly states that this is fire-and-forget telemetry and should not block the user flow.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local tools respira_get_site_context and respira_diagnose_connection to generate a status report. While this data comes from the user's WordPress environment, it is used for diagnostic reporting without triggering high-privilege automated actions. \n
  • Ingestion points: Site context data and connection status flags (SKILL.md). \n
  • Boundary markers: The output is structured within a specific 'Respira Connection Status' report section. \n
  • Capability inventory: The skill is diagnostic and does not include tools for arbitrary file writing or network requests based on the ingested data. \n
  • Sanitization: The ingested data is used for display in a predefined report format.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 03:53 AM