Skills
Navigate
skills/unbrowse-ai/sites/liftoff/Gen Agent Trust Hub

liftoff

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from liftoff.ai through search results and page reading tools, which could contain malicious instructions designed to influence the agent.
  • Ingestion points: liftoff_ai__get_search and liftoff_ai__read_page tools in SKILL.md ingest external web content.
  • Boundary markers: None identified in the skill instructions to distinguish between tool output and data content.
  • Capability inventory: The skill possesses network communication capabilities via the Unbrowse API and MCP transport.
  • Sanitization: No evidence of sanitization or filtering of the ingested web content before it is processed by the agent.
  • [EXTERNAL_DOWNLOADS]: The skill references and downloads configuration/code from the vendor's external infrastructure.
  • Evidence: References to https://unbrowse.ai/mcp/liftoff.ai for MCP transport and https://unbrowse.ai/api/v1/sites/liftoff.ai/openapi.json for API definitions.
  • [COMMAND_EXECUTION]: The documentation provides instructions for executing shell commands to install and configure the skill.
  • Evidence: README.md and SKILL.md contain commands for claude mcp add, codex mcp add, and npx skills add.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 10:56 AM