competitor-profiling
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingest data from untrusted external sources, which is a primary surface for indirect prompt injection attacks.
- Ingestion points: The skill uses
firecrawl_scrapeandfirecrawl_searchto fetch content from arbitrary competitor URLs and third-party review sites such as G2, Capterra, and Product Hunt (documented inSKILL.mdunder Phase 1 andreferences/tool-reference.md). - Boundary markers: The instructions lack specific boundary markers or "ignore embedded instructions" delimiters when directing the agent to synthesize scraped content into the final profile documents.
- Capability inventory: The agent has the capability to write markdown and JSON files to the local file system within the
competitor-profiles/directory and use specialized scraping and SEO tools (SKILL.md). - Sanitization: There are no documented sanitization steps or filtering mechanisms to prevent the agent from obeying malicious instructions that might be hidden in a competitor's website metadata, headlines, or user-submitted reviews.
Audit Metadata