The Agent Skills Directory

[No Code] (SAFE): The skill consists entirely of a markdown definition file (SKILL.md) with no accompanying executable code, scripts, or binary files. No malicious behavioral patterns were detected.- [Indirect Prompt Injection] (LOW): As a secret scanner, the skill is designed to ingest and process external, potentially untrusted data. This creates a surface for indirect prompt injection, although no specific vulnerable implementation logic is provided within the skill itself.
Ingestion points: Files or code blocks processed via the Read and Grep tools.
Boundary markers: Absent from the skill definition.
Capability inventory: Access to Read, Write, Grep, and Bash tools.
Sanitization: No evidence of sanitization or content filtering is provided in the skill instructions.

secret-scanner