skills/gdewitte/agentic/wayfinder/Gen Agent Trust Hub

wayfinder

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted data from an external issue tracker (ticket titles, descriptions, and comments) to determine the work frontier and make planning decisions. This creates a surface where malicious instructions embedded in tickets could influence the agent's logic or its use of sub-agents.
  • Ingestion points: The skill reads ticket titles, bodies, and metadata from the repository's issue tracker to build and update the map issue (SKILL.md).
  • Boundary markers: No explicit instructions or delimiters are used when processing ticket content to distinguish data from instructions.
  • Capability inventory: The agent can create, modify, and close tracker issues, and invoke multiple sub-agents including /research (which creates git branches) and /prototype (SKILL.md).
  • Sanitization: The skill does not implement sanitization or validation for input received from the issue tracker.
  • [COMMAND_EXECUTION]: The skill instructs the agent to invoke several external tools and sub-agents, such as /setup-matt-pocock-skills, /research, /prototype, /grilling, and /domain-modeling. These tools extend the agent's operational scope based on the content of the processed issues.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 12:30 PM
Security Audit — agent-trust-hub — wayfinder