sequential-thinking
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional, providing a mental framework for the AI agent to break down complex problems. It does not include any code, shell commands, file system access, or network operations.
- [INDIRECT_PROMPT_INJECTION]: The skill acts as a reasoning layer for processing user input. While it defines how input is decomposed, it does not provide the agent with any tools or capabilities that could be exploited by malicious data.
- Ingestion points: User requests are analyzed in the 'Decompose' step (Step 1).
- Boundary markers: The instructions recommend the use of
<thinking>blocks to isolate reasoning from final output. - Capability inventory: No tools, network access, or system commands are defined or used within the skill.
- Sanitization: Not applicable as the skill does not execute actions outside of its internal reasoning process.
Audit Metadata