scrapingbot

Pass

Audited by Gen Agent Trust Hub on May 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the @membranehq/cli package from the official NPM registry, which is a trusted tool from the skill's author.
  • [COMMAND_EXECUTION]: The skill uses the membrane CLI to perform authentication and manage connections to ScrapingBot. These are standard operations for the service.
  • [REMOTE_CODE_EXECUTION]: The skill executes actions on the remote Membrane platform via the CLI. This is the core functionality of the tool and is managed by the service provider.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes scraped data. 1. Ingestion points: Data returned from ScrapingBot API calls (SKILL.md). 2. Boundary markers: Absent; no specific instructions are provided to separate scraped content from agent instructions. 3. Capability inventory: The agent can execute membrane action run, which can interact with other connected services. 4. Sanitization: No explicit sanitization or filtering is described for the incoming web data.
  • [CREDENTIALS_UNSAFE]: The skill follows security best practices by directing the agent to use Membrane's connection system instead of requesting raw API tokens from the user.
Audit Metadata
Risk Level
SAFE
Analyzed
May 2, 2026, 02:49 PM
Security Audit — agent-trust-hub — scrapingbot