dalio-principles-allweather
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external financial, market, and structural data for investment analysis. While this represents a potential attack surface for indirect prompt injection, the skill lacks tool access or execution capabilities to perform harmful actions.
- Ingestion points: The skill accepts financial, market, regime, or structural data as input (referenced in SKILL.md and invest.md).
- Boundary markers: No specific delimiters or boundary instructions are used to separate untrusted data from instructions.
- Capability inventory: The skill contains no subprocess calls, file-write operations, or network tools.
- Sanitization: No sanitization or validation of the input data is described.
- [NO_CODE]: The skill consists entirely of markdown documentation and configuration metadata. It does not contain any executable scripts, shell commands, or binary files.
Audit Metadata