aikey-resonate
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided content which serves as an entry point for indirect prompt injection attacks.
- Ingestion points: In
SKILL.md, the skill defines three modes (A, B, and C) that ingest user drafts, external viral post content, and descriptions of phenomena for analysis. - Boundary markers: The instructions do not utilize specific delimiters or isolation techniques (such as XML tags or strictly defined blocks) to separate user-provided content from the agent's internal diagnostic logic.
- Capability inventory: The skill is capable of reading and writing to the
{vault}directory for maintaining 'skill memory' and referencing theory bases. It does not have access to general shell commands or arbitrary network requests. - Sanitization: No explicit sanitization, filtering, or escaping mechanisms for the ingested text are described in the instructions.
Audit Metadata