wayfinder
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from a repository's issue tracker, including issue bodies, titles, and comments, which are used to orient the agent and resolve decisions. This content is inherently untrusted and could be influenced by malicious actors with access to the tracker. The instructions lack explicit boundary markers or sanitization logic to distinguish between the map's structure and potentially malicious instructions embedded in the tickets.
- Ingestion points: Issue bodies, titles, and comments from the tracker as described in the 'The Map' and 'Tickets' sections of SKILL.md.
- Boundary markers: Absent; the skill does not define specific delimiters or instructions to ignore embedded commands within the fetched issue content.
- Capability inventory: The agent has the capability to modify issues, close tickets, run subagents for research, and call specialized tools for prototyping.
- Sanitization: Absent; no explicit sanitization or validation of the ingested issue content is described.
Audit Metadata