wayfinder

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill logic is confined to document organization within a designated local directory ('docs/scratch/'). It lacks any functionality for system modification or network exfiltration.
  • [NO_CODE]: The skill consists entirely of instructional Markdown and configuration files, containing no executable scripts or binary files.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process content from various project files and research notes, which presents an inherent surface for indirect prompt injection. However, since the skill has no access to dangerous tools (like shell execution or network calls), this surface cannot be effectively exploited to cause harm. (1) Ingestion points: WAYFINDER.md, decision tickets in docs/scratch/wayfinder/, and various project files read during research phases. (2) Boundary markers: None implemented in the skill instructions. (3) Capability inventory: Reading and writing local Markdown files. (4) Sanitization: No sanitization of input data is performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 02:54 AM
Security Audit — agent-trust-hub — wayfinder