sequential-thinking

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Dynamic Execution (SAFE): The skill invokes a local script (sequential-thinking.ts) using the Bun runtime. This is the intended core functionality and does not execute untrusted or remote code.
  • Indirect Prompt Injection (SAFE): The skill ingests text through the --thought flag.
  • Ingestion points: Textual thoughts are passed as command-line arguments as seen in the SKILL.md usage examples.
  • Boundary markers: None identified in the provided script invocation examples.
  • Capability inventory: The skill performs local script execution via the bun runtime environment.
  • Sanitization: While no explicit sanitization is documented for the --thought input, the primary use case is internal agent reasoning, which minimizes the risk of adversarial input causing harm.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:30 PM