profile-model
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted external data, creating a surface for indirect prompt injection.
- Ingestion points: The agent is instructed to load user-specified .json or .json.gz profiler traces into a pandas DataFrame (SKILL.md, Step 2).
- Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions that might be embedded within the trace file's data fields.
- Capability inventory: The skill uses the loaded data to perform performance analysis via pandas queries and the HTA library's diagnostic functions (SKILL.md, Steps 4 & 6).
- Sanitization: While the skill manages data types for storage, it does not perform semantic sanitization of trace fields like 'name' or 'args' before including them in the final analysis output provided to the user.
Audit Metadata