wayfinder
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest data from
AGENTS.local.mdfiles and external issue trackers, providing a surface for malicious instructions to influence agent behavior.\n - Ingestion points: The instructions mandate finding and reading all
AGENTS.local.mdfiles in the repository before any other action and querying issue trackers for map and ticket details.\n - Boundary markers: The skill does not include delimiters or specific instructions to ignore embedded prompts within the external data sources.\n
- Capability inventory: The agent can write to issue trackers, create git branches, and invoke subagents or platform tools for research and prototyping.\n
- Sanitization: No validation, filtering, or sanitization processes are described for the content retrieved from the repository or the tracker.
Audit Metadata