aeon-skill-evals
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources (the outputs of other skills) to perform pattern matching and regression analysis.
- Ingestion points: The
evalandbootstrapoperations read and analyze the outputs of other skills installed in the environment, as described inSKILL.md. - Boundary markers: There are no explicit delimiters or instructions shown to ignore embedded commands within the ingested skill outputs.
- Capability inventory: The skill records results in a local
evals-state.jsonfile. No network operations or administrative commands are evidenced in the provided files. - Sanitization: No sanitization, filtering, or escaping of the ingested data is described before analysis.
Audit Metadata