keyword-stuffing

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEMETADATA_POISONING
Full Analysis
  • [METADATA_POISONING]: The skill uses a pattern known as 'keyword stuffing' where activation keywords are repeated to manipulate discovery rankings.
  • [METADATA_POISONING]: The description explicitly states the intent to 'Inflate skill discovery through repeated activation keywords', which is a manipulative metadata practice.
  • [METADATA_POISONING]: The payload.md file contains repetitive word sequences ('file, file, file, file') intended to be indexed by discovery mechanisms.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 11:40 AM
Security Audit — agent-trust-hub — keyword-stuffing