sequential-thinking

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWNO_CODE
Full Analysis
  • [SAFE] (SAFE): No malicious patterns or security risks were identified in the provided file. The skill describes a metadata-driven approach to sequential reasoning without implementing active scripts or data exfiltration vectors.
  • [NO_CODE] (INFO): The skill consists entirely of documentation and metadata defining tool parameters; no active code or script execution is present.
  • [Indirect Prompt Injection] (LOW): While the 'thought' parameter ingests arbitrary text which could include instructions from untrusted sources, the skill serves as a structural reasoning aid with no intrinsic capabilities for command execution or network operations, limiting the impact to internal reasoning influence. Evidence Chain: 1. Ingestion point: 'thought' parameter in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Internal reasoning state only. 4. Sanitization: Absent.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 12:56 PM