fiverr-research
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local bash helper script (
scripts/crawlora.sh) to interact with the Crawlora API. The script features robust security controls, including protecting the API key from command-line disclosure via temporary curl configuration files and using strict route whitelisting to restrict API access to documented endpoints. - [EXTERNAL_DOWNLOADS]: The skill performs authenticated requests to the vendor's API at
api.crawlora.netto fetch Fiverr gig and seller data. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Fiverr gig listings and seller profiles, representing a standard indirect prompt injection surface.
- Ingestion points: External data enters the agent's context through API responses processed in
scripts/crawlora.sh. - Boundary markers: The skill does not explicitly provide instructions or delimiters to isolate untrusted external content from the agent's primary instructions.
- Capability inventory: The skill possesses capabilities for network operations to a specific vendor API and execution of a local script.
- Sanitization: While the script validates input parameters, the retrieved content from Fiverr is passed as-is in JSON format without specific sanitization for embedded prompt instructions.
Audit Metadata