external-gitcode-ascend-ascend-profiling-anomaly

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security risks were detected. The skill is dedicated to NPU performance profiling analysis.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted profiling data (CSV, JSON) which could theoretically contain malicious instructions. However, the analysis logic focuses on quantitative metrics and specific kernel names, and the skill lacks dangerous capabilities like network access, minimizing the risk. 1. Ingestion points: kernel_details.csv, trace_view.json, op_summary, communication.json (described in SKILL.md and references/kernel_data_guide.md). 2. Boundary markers: Absent. 3. Capability inventory: File reading and report generation (Markdown). 4. Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 03:04 AM
Security Audit — agent-trust-hub — external-gitcode-ascend-ascend-profiling-anomaly