apify-influencer-brand-collabs

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Instagram (profiles, posts, and reels), which represents a vulnerability surface for indirect prompt injection attacks where malicious instructions could be embedded in scraped content.
  • Ingestion points: The skill ingests data from four Apify actors (instagram-profile-scraper, brand-collaboration-scraper, instagram-post-scraper, and instagram-reel-scraper) as outlined in the pipeline in SKILL.md.
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the scraped data as untrusted or to ignore instructions found within the data.
  • Capability inventory: The skill allows the agent to call additional tools and perform data aggregation based on the inputs received from external sources.
  • Sanitization: No sanitization, validation, or filtering of the retrieved text data is performed before it enters the agent's context.
  • [EXTERNAL_DOWNLOADS]: The skill relies on external Apify actors to perform its primary functions. These are legitimate tools provided by the skill's author ('apify') and are invoked via standard platform MCP tools.
  • Evidence: The skill instructs the agent to use mcp__claude_ai_Apify__call-actor to execute apify/instagram-profile-scraper, apify/brand-collaboration-scraper, apify/instagram-post-scraper, and apify/instagram-reel-scraper.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 04:15 PM
Security Audit — agent-trust-hub — apify-influencer-brand-collabs