think-random-frameworks
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied topics and 'stuck framings' which are interpolated into thinking framework templates. While it lacks sanitization, the risk is inherent to the thinking process and mitigated by the skill's structure.
- Ingestion points: The skill takes a user's situation or problem description as input in
SKILL.md(Step 1) andreferences/engine.md(Step 0). - Boundary markers: The output template in
references/TEMPLATE.mduses clear headers and markdown to separate the user's topic from the framework analysis, though it does not use explicit 'ignore instructions' markers. - Capability inventory: The skill only performs text generation and analysis; it has no file-write, network, or subprocess capabilities.
- Sanitization: No explicit sanitization of the input text is performed before it is used to populate framework artifacts.
Audit Metadata