dspy-prompting
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS] (SAFE): The skill suggests installing the 'dspy' Python package. This is a legitimate library from a reputable academic source (Stanford NLP).
- [PROMPT_INJECTION] (LOW): The skill demonstrates patterns for processing untrusted data, which constitutes an indirect prompt injection surface. 1. Ingestion points: Untrusted data enters via 'dspy.InputField()' in classes like 'QA', 'MathProblem', 'SearchQA', and 'ExtractPerson' in 'SKILL.md'. 2. Boundary markers: No delimiters or specific 'ignore instructions' warnings are present in the examples. 3. Capability inventory: The skill utilizes 'dspy.Predict', 'dspy.ChainOfThought', and 'dspy.ReAct' for reasoning and tool execution. 4. Sanitization: No sanitization or validation of the input fields is shown in the provided code snippets.
Audit Metadata