wayfinder
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from an external issue tracker (map and ticket bodies). This creates an attack surface for indirect prompt injection where malicious instructions embedded in tickets could influence the agent's actions or tool usage.\n
- Ingestion points: The agent reads map metadata, 'Notes', and ticket descriptions from a repository's issue tracker (e.g., GitHub or GitLab).\n
- Boundary markers: Absent; the instructions do not specify delimiters to separate untrusted issue content from system instructions.\n
- Capability inventory: The skill can modify issue states, create git branches, and trigger other sub-skills such as
/research,/prototype,/grilling, and/domain-modeling.\n - Sanitization: No explicit validation or filtering of the issue tracker content is mentioned before it is used to orient the agent's session.\n- [COMMAND_EXECUTION]: The instructions direct the agent to execute specific setup commands like
/setup-matt-pocock-skillsto configure the environment. While this is likely an initialization tool for the platform or community framework, it involves the execution of external logic to define the working environment.
Audit Metadata