wayfinder
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is purely instructional, composed of Markdown and YAML configuration. It contains no scripts, binaries, or command-line executions. All external references are for attribution purposes to a known repository.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that involves reading and processing content from external issue trackers, which serves as an ingestion surface for untrusted data.
- Ingestion points: The agent is instructed to fetch the 'map' issue and child ticket bodies from a repository's issue tracker (SKILL.md).
- Boundary markers: The instructions mandate specific Markdown structures (e.g., ## Destination, ## Question) to organize data, which helps provide context but does not fully prevent adversarial injection.
- Capability inventory: The skill uses the agent's platform tools to perform tracker operations; it does not ship its own implementation of file-write or network-send capabilities beyond what the platform provides.
- Sanitization: No explicit sanitization or filtering logic is provided within the skill instructions, relying on the underlying agent platform's handling of external data.
Audit Metadata