tenders-finder

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple third-party tender portals, creating a surface for indirect prompt injection if a portal contains malicious instructions.
  • Ingestion points: The skill uses tinyfish agent run to scrape data from gebiz.gov.sg, tendersontime.com, biddetail.com, tendersinfo.com, and globaltenders.com (SKILL.md).
  • Boundary markers: The agent prompts include specific extraction tasks and a requested JSON return format, which acts as a structural boundary, though it does not explicitly instruct the agent to ignore instructions embedded in the scraped data.
  • Capability inventory: The skill identifies and displays tender information; it does not possess capabilities for file writing, system modification, or subsequent network operations based on the ingested data.
  • Sanitization: There is no evidence of sanitization or filtering of the scraped content (e.g., tender titles or descriptions) before it is presented to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 01:04 AM
Security Audit — agent-trust-hub — tenders-finder