The Agent Skills Directory

[SAFE] (SAFE): The skill's behavior is consistent with its stated purpose of performance auditing. It uses read-only operations on user-provided data and does not exhibit patterns of data exfiltration or persistence.
[Indirect Prompt Injection] (LOW): The skill processes untrusted trace data which represents a potential surface for indirect prompt injection. However, the risk is minimal due to the restricted toolset and the lack of outbound network or file-write capabilities.
Ingestion points: User-provided trace JSON files via Read and Grep tools.
Boundary markers: Absent; the skill does not explicitly instruct the agent to treat data as untrusted or ignore instructions within the JSON.
Capability inventory: Read, Grep (No network, no file-write, no shell execution).
Sanitization: Absent; the skill extracts values directly from the JSON fields for reporting.

trace-audit